Wireless Networking in the Developing World

Second Edition

A practical guide to planning and building low-cost telecommunications infrastructure

Wireless Networking in the Developing World
For more information about this project, visit us online at http://wndw.net/

First edition, January 2006 Second edition, December 2007

Many designations used by manufacturers and vendors to distinguish their products are claimed as trademarks. Where those designations appear in this book, and the authors were aware of a trademark claim, the designations have been printed in all caps or initial caps. All other trademarks are property of their respective owners. The authors and publisher have taken due care in preparation of this book, but make no expressed or implied warranty of any kind and assume no responsibility for errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of the use of the information contained herein.

© 2007 Hacker Friendly LLC, http://hackerfriendly.com/

This work is released under the Creative Commons Attribution-ShareAlike 3.0 license. For more details regarding your rights to use and redistribute this work, see http://creativecommons.org/licenses/by-sa/3.0/

Contents
Where to Begin 1
Purpose of this book ........................................................................................................................... 2 Fitting wireless into your existing network.......................................................................................... 3 Wireless networking protocols............................................................................................................. 3 Question & Answer ............................................................................................................................. 5

A Practical Introduction to Radio Physics

9

What is a wave?................................................................................................................................. 9 Polarization...................................................................................................................................... 13 The electromagnetic spectrum........................................................................................................... 13 Bandwidth........................................................................................................................................ 15 Frequencies and channels................................................................................................................. 15 Behavior of radio waves................................................................................................................... 15 Line of sight...................................................................................................................................... 22 Power............................................................................................................................................... 24 Physics in the real world ................................................................................................................... 26

Network Design

27

Designing the physical network......................................................................................................... 51 802.11 wireless networks.................................................................................................................. 54 Mesh networking with OLSR.............................................................................................................. 56 Estimating capacity ........................................................................................................................... 65 Traffic optimization........................................................................................................................... 79 Internet link optimization................................................................................................................. 89 More information..............................................................................................................................93

Antennas & Transmission Lines

95

Cables.............................................................................................................................................. 95 Waveguides...................................................................................................................................... 97 Connectors and adapters .................................................................................................................100 Antennas & radiation patterns........................................................................................................ 102 Reflector theory.............................................................................................................................. 114 Amplifiers....................................................................................................................................... 115 Practical antenna designs................................................................................................................116

Networking Hardware

135

Wired wireless ................................................................................................................................ 135 Choosing wireless components........................................................................................................ 137 Commercial vs. DIY solutions........................................................................................................... 139 Building an access point from a PC ..................................................................................................143

Security & Monitoring

157

Physical security ............................................................................................................................. 158 Threats to the network .................................................................................................................... 160 Authentication................................................................................................................................ 162 Privacy ........................................................................................................................................... 167 Network Monitoring........................................................................................................................ 174 What is normal? ............................................................................................................................. 203

Solar Power

211

Solar energy................................................................................................................................... 211 Photovoltaic system components..................................................................................................... 212 The solar panel............................................................................................................................... 217 The battery..................................................................................................................................... 222 The power charge regulator ............................................................................................................ 229 Converters...................................................................................................................................... 231 Equipment or load.......................................................................................................................... 232 How to size your photovoltaic system.............................................................................................. 238 Cost of a solar installation...............................................................................................................246

Building an Outdoor Node

249

Waterproof enclosures.................................................................................................................... 249 Providing power..............................................................................................................................250 Mounting considerations................................................................................................................. 251 Safety.............................................................................................................................................257 Aligning antennas on a long distance link....................................................................................... 258 Surge and lightning protection........................................................................................................ 263

Troubleshooting

267

Building your team......................................................................................................................... 267 Proper troubleshooting technique................................................................................................... 270 Common network problems ............................................................................................................ 271

Economic Sustainability

281

Create a Mission Statement............................................................................................................. 282 Evaluate the Demand for Potential Offerings ...................................................................................283 Establish Appropriate Incentives...................................................................................................... 284 Research the Regulatory Environment for Wireless.......................................................................... 286 Analyze the Competition................................................................................................................. 286 Determine Initial and Recurring Costs and Pricing........................................................................... 287 Secure the Financing....................................................................................................................... 291 Evaluate the Strengths and Weaknesses of the Internal Situation..................................................... 293 Putting it All Together..................................................................................................................... 294 Conclusion...................................................................................................................................... 297

Case Studies

299

General advice ................................................................................................................................299 Case study: Crossing the divide with a simple bridge in Timbuktu.................................................... 302 Case study: Finding solid ground in Gao.......................................................................................... 305 Case Study: Fantsuam Foundation's Community Wireless Network................................................... 308 Case study: The quest for affordable Internet in rural Mali.............................................................. 319 Case study: Commercial deployments in East Africa......................................................................... 325 Case study: Dharamsala Community Wireless Mesh Network ........................................................... 332 Case study: Networking Mérida State .............................................................................................. 334 Case study: Chilesincables.org......................................................................................................... 345 Case study: Long Distance 802.11.................................................................................................... 355

Appendix A: Resources Appendix B: Channel Allocations Appendix C: Path Loss Appendix D: Cable Sizes Appendix E: Solar Dimensioning

371 379 381 382 383

About This Book
This book is part of a set of related materials about the same topic: Wireless Networking in the Developing World. The WNDW project includes: • Printed books, available on demand • Several translations, including French, Spanish, Portuguese, Italian, Arabic, and others • A DRM-free PDF and HTML version of the book • An archived mailing list for discussion of the concepts and techniques described in the book • Additional case studies, training course material, and related information For all of this material and more, see our website at http://wndw.net/ The book and PDF file are published under a Creative Commons AttributionShareAlike 3.0 license. This allows anyone to make copies, and even sell them for a profit, as long as proper attribution is given to the authors and any derivative works are made available under the same terms. Any copies or derivative works must include a prominent link to our website, http://wndw.net/. See http://creativecommons.org/licenses/by-sa/3.0/ for more information about these terms. Printed copies may be ordered from Lulu.com, a print-on-demand service. Consult the website (http://wndw.net/) for details on ordering a printed copy. The PDF will be updated periodically, and ordering from the print-ondemand service ensures that you will always receive the latest revision. The website will include additional case studies, currently available equipment, and more external website references. Volunteers and ideas are welcome. Please join the mailing list and send ideas. The training course material was written for courses given by the Association for Progressive Communications and the Abdus Salam International Center for Theoretical Physics. See http://www.apc.org/wireless/ and http://wireless.ictp.trieste.it/ for more details on those courses and their material. Additional information was provided by the International Network for the Availability of Scientific Publications, http://www.inasp.info/. Some of this material has been incorporated directly into this book. Additional material was adapted from How To Accelerate Your Internet, http://bwmo.net/.

Credits
This book was started as the BookSprint project at the 2005 session of WSFII, in London, England (http://www.wsfii.org/). A core team of seven people built the initial outline over the course of the event, presented the results at the conference, and wrote the book over the course of a few months. Throughout the project, the core group has actively solicited contributions and feedback from the wireless networking community. Add your own feedback and updates to the WNDW wiki at http://wiki.wndw.net/. • Rob Flickenger was the lead author and editor of this book. Rob has written and edited several books about wireless networking and Linux, including Wireless Hacks (O Reilly Media) and How To Accelerate Your Internet (http://bwmo.net/). He is proud to be a hacker, amateur mad scientist, and proponent of free networks everywhere. • Corinna “Elektra” Aichele. Elektra s main interests include autonomous power systems and wireless communication (antennas, wireless long shots, mesh networking). She made a small linux distro based on slackware geared to wireless mesh networking. This information is of course redundant if one reads the book... http://www.scii.nl/~elektra • Sebastian Büttrich (http://wire.less.dk/) is a generalist in technology with a background in scientific programming and physics. Originally from Berlin, Germany, he worked with IconMedialab in Copenhagen from 1997 until 2002. He holds a Ph.D. in quantum physics from the Technical University of Berlin. His physics background includes fields like RF and microwave spectroscopy, photovoltaic systems, and advanced maths. He is also a performing and recording musician. • Laura M. Drewett is a Co-Founder of Adapted Consulting Inc., a social enterprise that specializes in adapting technology and business solutions for the developing world. Since Laura first lived in Mali in the 1990s and wrote her thesis on girls education programs, she has strived to find sustainable solutions for development. An expert in sustainability for ICT projects in developing world environments, she has designed and managed projects for a diversity of clients in Africa, the Middle East and Eastern Europe. Laura holds a Bachelors of Arts with Distinction in Foreign Affairs and French from the University of Virginia and a Master s Certificate in Project Management from the George Washington University School of Business. • Alberto Escudero-Pascual and Louise Berthilson are the founders of IT +46, a Swedish consultancy company with focus on information technology in developing regions. IT +46 is internationally known for promoting and implementing wireless Internet infrastructure in rural areas of Africa and Latinoamerica. Since 2004, the company has trained over 350 people in 14

countries and released over 600 pages of documentation under Creative Commons License. More information can be found at http://www.it46.se/ • Carlo Fonda is a member of the Radio Communications Unit at the Abdus Salam International Center for Theoretical Physics in Trieste, Italy. • Jim Forster has spent his career in software development, mostly working on operating systems and networking in product companies. He has experience with several failed startup companies in Silicon Valley, and one successful one, Cisco Systems. After a lot of product development work there, his more recent activities involve projects and policies for improving Internet access in developing countries. He can be reached at jrforster@mac.com. • Ian Howard. After flying around the world for seven years as a paratrooper in the Canadian military, Ian Howard decided to trade his gun for a computer. After finishing a degree in environmental sciences at the University of Waterloo he wrote in a proposal, "Wireless technology has the opportunity to bridge the digital divide. Poor nations, who do not have the infrastructure for interconnectivity as we do, will now be able to create a wireless infrastructure." As a reward, Geekcorps sent him to Mali as the Geekcorps Mali Program Manager, where he led a team equipping radio stations with wireless interconnections and designed content sharing systems. He is now a consultant on various Geekcorps programs. • Kyle Johnston, http://www.schoolnet.na/ • Tomas Krag spends his days working with wire.less.dk, a registered nonprofit, based in Copenhagen, which he founded with his friend and colleague Sebastian Büttrich in early 2002. wire.less.dk specialises in community wireless networking solutions, and has a special focus on low-cost wireless networks for the developing world. Tomas is also an associate of the Tactical Technology Collective http://www.tacticaltech.org/, an Amsterdam-based non-profit “to strengthen social technology movements and networks in developing and transition countries, as well as promote civil society s effective, conscious and creative use of new technologies.” Currently most of his energy goes into the Wireless Roadshow (http://www.thewirelessroadshow.org/), a project that supports civil society partners in the developing world in planning, building and sustaining connectivity solutions based on license-exempt spectrum, open technology and open knowledge. • Gina Kupfermann is graduate engineer in energy management and holds a degree in engineering and business. Besides her profession as financial controller she has worked for various self-organised community projects and nonprofit organisations. Since 2005 she is member of the executive board of the development association for free networks, the legal entity of freifunk.net.

He now resides in Amman. a campaign for spreading knowledge and social networking about free and open networks. Messer worked for 11 years in eastern and southern Africa in voice and data communications for startups and multinational cellular carriers. He began his ICT career in the early 1990s with a bulletin board system (BBS) on an analog modem and has since continued to create systems that enhance communication. and he is happy to have merged the two together working in the field of wireless networking.com/) was the lead copy editor. Adam Messer metamorphosed into a telecommunications professional after a chance conversation in 1995 led him to start one of Africa's first ISPs. In 2002 he cofounded www. Inc. school computer labs and lighting systems for rural communities. In this capacity. • Juergen Neumann (http://www.freifunk. • Catherine Sharp (http://odessablue. • Casey Halverson (http://seattlewireless. . • Ermanno Pietrosemoli has been involved in planning and building computer networks for the last twenty years. Most recently. Frédéric spent more than a year at IESC/Geekcorps Mali as a consultant. • Richard Lotz (http://greenbits. • Marco Zennaro. Support • Lisa Chan (http://www. he has been teaching wireless data communications in several countries while keeping his base at Mérida. Freifunk is globally regarded as one of the most successful community-projects in this field. www.• Adam Messer. • Frédéric Renet is a co-founder of Technical Solutions at Adapted Consulting. As president of the Latin American Networking School.ergomedia. As a consultant for ICT strategy and implementation. Jordan. aka marcusgennaroz.net/~casey/) provided technical review and suggestions. He has been using BBSes and ham radios since he was a teenager.com/) provided several updated illustrations for this edition.com/) provided copy edit support.cowinanorange. Frédéric has been involved in ICT for more than 10 years and has worked with computers since his childhood. Venezuela. Escuela Latinoamericana de Redes “EsLaRed”. he designed many innovative solutions for FM radio broadcasting. Italy. He still carries his Apple Newton. • Jessie Heaven Lotz (http://jessieheavenlotz.eslared.net/~rlotz/) provided technical review and suggestions. He works on SeattleWireless projects and would like to take his node (and his house) off the grid.org.ve. is an electronic engineer working at the ICTP in Trieste.net. he has worked for major German and international companies and many non-profit projects. Originally trained as an insect scientist. Pioneering wireless data services in Tanzania.de/) started working with information technology in 1984 and since then has been looking for ways to deploy ICT in useful ways for organizations and society.

net/) and is an evangelist for FreeNetworks worldwide. WA. About the solar power guide The source material for the Solar Power chapter was translated and developed by Alberto Escudero-Pascual.net/~mattw/) provided technical review and copy edit support. The publication of this work has been supported by Canada s International Development Research Centre. the organization Engineering without Borders (Spanish Federation) published the first version of a handbook titled "Manual de Energía Solar Fotovoltaica y Cooperación al Desarrollo".0. In 1998. This new work is licensed under Creative Commons Attribution-ShareAlike 3.• Lara Sobel designed the cover for WNDW 2nd Edition. We would especially like to thank community networkers everywhere. She is an artist currently living in Seattle. Additional support was provided by NetworktheWorld. none of the members of the editorial team kept the document in electronic format and more editions were never made. community networks could not exist.idrc. Mercedes Montero Bartolomé y Julio Amador. and occasional bandwidth that served as the incubator for this project. We hope that this material becomes a new departure point for new editions including new contributions by the community. . Without you. They have passed almost ten years from that very first edition and this document is an effort to rescue and to extend the handbook. http://www. The handbook was written and published by members of the NGO and experts of the Institute of Energy Solar of the Polytechnical University of Madrid. Special thanks The core team would like to thank the organizers of WSFII for providing the space. As part of this rescue operation Alberto would like to thank the coordinators of the first original edition and his mentors in his years at University: Miguel Ángel Eguido Aguilera.ca/. support. Matt is the founder of SeattleWireless (http://seattlewireless. who devote so much of their time and energy towards fulfilling the promise of the global Internet. By curiosities of life.org. This second and extended edition of the solar power guide has received valuable input from Frédéric Renet and Louise Berthilson. • Matt Westervelt (http://seattlewireless.

People all over the world are finding that Internet access gives them a voice to discuss their problems. people can finally begin to have a stake in building their own communications infrastructure. in a way that the telephone and television simply cannot compete with. But building wireless networks is only partly about saving money. Communities connected to the Internet at high speed have a voice in a global marketplace. but also show how we have done it. The time and effort saved by having access to the global network of information translates into wealth on a local scale.1 Where to Begin This book was created by a team of individuals who each. 1 . We believe that by taking advantage of this state of affairs. and whatever else is important to their lives. are actively participating in the ever-expanding Internet by pushing its reach farther than ever before. We hope to not only convince you that this is possible. while equipment capabilities continue to sharply increase. where transactions happen around the world at the speed of light. Wireless infrastructure can be built for very little cost compared to traditional wired alternatives. and that reality is being built on wireless networks. as more work can be done in less time and with less effort. the network becomes all the more valuable as more people are connected to it. in their own field. they will directly benefit from what the Internet has to offer. and to give you the information and tools you need to start a network project in your local community. By providing people in your local community with cheaper and easier access to information. The massive popularity of wireless networking has caused equipment costs to continually plummet. What has until recently sounded like science fiction is now becoming a reality. Likewise. politics.

information is presented from many points of view. these techniques are intended to augment existing systems. the networks described in this book are data networks. CDMA. The emphasis is on building infrastructure links intended to be used as the backbone for wide area wireless networks. knowledge and trust are spread throughout the community. since the cost of deploying these technologies is well beyond the reach of most community projects. We specifically do not cover GSM. voice. and the ultimate results of these attempts. By getting local people involved in the construction of the network. They allow people to collaborate on projects across wide distances. provide broadband network access in areas that even dialup does not exist. email. and financial factors. Since the first spark gap experiments at the turn of the last century. and people begin to understand the importance of having a share in their communications infrastructure. While such a network can carry data. And by working with your local community. and ultimately connect you and your neighbors to the global Internet. In this book we will focus on wireless data networking technologies in the 802. By using local sources for materials and fabricating parts yourself. and video (as well as traditional web and Internet traffic). you can build high speed data networks that connect remote areas together.11 family. Voice communications. This book is not a guide to configuring a radio card in your laptop or choosing consumer grade gear for your home network. and provide connectivity in areas where running fiber or other physical cable would be impractical. wireless has been a rapidly evolving area of communications technology.2 Chapter 1: Where to Begin But even without access to the Internet. the techniques described in this book are not intended to replace existing wired infrastructure (such as telephone systems or fiber optic backbone). they realize that communication networks are built to allow people to connect with each other. or other wireless voice technologies. you can build a telecommunications infrastructure that benefits everyone who participates in it. The extensive collection of case studies present various groups attempts at building these networks. the resources that were committed to them. including technical. and other data can be exchanged for very little cost. . Rather. Ultimately. While we provide specific examples of how to build working high speed data links. wireless community networks have tremendous value. you can build reliable network links with very little budget. Purpose of this book The overall goal of this book is to help you build affordable communication technology in your local community by making best use of whatever resources are available. social. Using inexpensive off-the-shelf equipment. With that goal in mind.

11 family of radio protocols (802.11b. also known in many circles as Wi-Fi. from a simple extension (like a several kilometer Ethernet cable) to a distribution point (like a large hub). By implementing a common set of protocols.11g) have enjoyed an incredible popularity in the United States and Europe. Fitting wireless into your existing network If you are a network administrator.11 family of protocols. If manufacturers had chosen to implement their .11a.1: Some wireless networking examples. you may wonder how wireless might fit into your existing network infrastructure. As a result. Here just a few examples of how your network can benefit from wireless technology.Chapter 1: Where to Begin 3 We hope you find this book useful for solving your communication challenges. manufacturers world wide have built highly interoperable equipment. consumers are able to purchase low-cost equipment at a volume which has benefitted manufacturers. Wireless can serve in many capacities. This decision has proven to be a significant boon to the industry and the consumer. Wireless networking protocols The primary technology used for building low-cost wireless networks is currently the 802. Consumers are able to use equipment that implements 802. and 802. Internet University 10+ Km wireless link Remote office Campus network Wireless client Firewall Access point Wireless client Wireless client Wireless mesh node Wireless mesh node Wireless mesh node Wireless mesh node Wireless mesh node Wireless mesh node Wireless mesh node Internet Figure 1. The 802.11 without fear of “vendor lock-in”. 802.

New equipment and standards (such as . and increased range.16 (also known as WiMax) will likely solve some difficult problems currently observed with 802. It has a maximum data rate of 54 Mbps (with usable throughput of about 22 Mbps). This makes it incompatible with 802. Check with your local authorities before using 802.11g is a relative latecomer to the wireless marketplace. It has a maximum data rate of 54 Mbps.11g uses the same ISM range as 802. As it wasn t finalized until June 2003. 802. 802. Ratified by the IEEE on September 16. they have a long way to go to match the popularity and price point of 802. it is unlikely that wireless networking would be as inexpensive and ubiquitous as it is today. While this portion of the spectrum is relatively unused compared to 2.11a equipment.11a. 802.495 GHz. and not all are directly related to the radio protocol itself.11g is now the de facto standard wireless networking protocol as it now ships as a standard feature on virtually all laptops and most handheld devices.805 GHz. As equipment that supports WiMax is just becoming available at the time of this writing. • 802. with actual usable data speeds up to about 5 Mbps. and the higher frequency means shorter range compared to 802.11. There are many protocols in the 802. 802.11b. Despite the late start. and in a portion of the UNII band between 5. 1999.11a uses OFDM. and purchasing them will effectively lock you into that vendor for every part of your network.745 and 5.11b.11b.11a equipment is still quite inexpensive.11b/g at the same power. 802. Unfortunately these extensions will not operate between equipment from different manufacturers. we will focus primarily on the 802. there are a number of vendor-specific extensions to equipment.4 GHz. touting higher speeds. and can fall back to 11 Mbps DSSS or slower for backwards compatibility with the hugely popular 802.4 Chapter 1: Where to Begin own proprietary protocols.11 family.11b/g.11b is probably the most popular wireless networking protocol in use today. It uses a modulation called Direct Sequence Spread Spectrum (DSSS) in a portion of the ISM band from 2.150 and 5. 802. 1999.11a operates in the ISM band between 5.11g. Millions of devices supporting it have shipped since 1999. 802.320 GHz. Also ratified by the IEEE on September 16. It has a maximum rate of 11 Mbps. it is unfortunately only legal for use in a few parts of the world. but is not nearly as popular as 802. with actual throughput of up to 27 Mbps. In addition to the above standards.400 to 2.11 family. stronger encryption. The three wireless standards currently implemented in most readily available gear are: • 802. but uses a modulation scheme called Orthogonal Frequency Division Multiplexing (OFDM). • 802.11 equipment.11b or 802.11g. particularly in outdoor applications. While new protocols such as 802.

Due to the ubiquity of equipment and unlicensed nature of the 2. 802. this book will concentrate on building networks using 802.Chapter 1: Where to Begin 5 802. 802.11n.16. but this equipment is just starting to ship at the time of this writing. 267 • What are the most common problems encountered on wireless networks. if there is no grid power available? Page 211 • Do I need to run a power cable all the way up the tower? Page 250 • How can I use solar panel to power my wireless node while keeping it online overnight? Page 217 • How long will my access point run on a battery? Page 238 • Can I use a wind generator to power my equipment at night? Page 212 Management • How much bandwidth will I need to purchase for my users? Page 65 • How can I monitor and manage remote access points from my office? Page 174 • What do I do when the network breaks? Page 174. MIMO and WiMAX) promise significant increases in speed and reliability. Power • How can I supply power to my radio equipment.11y.11g. Question & Answer If you are new to wireless networking.4 GHz ISM band.11b and 802. with answers and suggestions on the listed page. you likely have a number of questions about what the technology can do and what it will cost. and availability and vendor interoperability is still uncertain. Here are some commonly asked questions. and how do I fix them? Page 267 Distance • How good is the range of my access point? Page 67 • Is there any formula I can use to know how far I can go with a given access point? Page 67 • How can I know if a remote place can be connected to Internet using a wireless link? Page 67 • Is there any software that can help me estimate the feasibility of a long distance wireless link? Page 74 .

6 Chapter 1: Where to Begin • The manufacturer says my access point has a range of 300 meters. or can I go without them? Page 263 • Can I build an antenna mast by myself? How high can I go? Page 251 • Why does my antenna work much better when I mount it “sideways”? Page 13 • Which channel should I use? Page 15 • Will radio waves travel through buildings and trees? What about people? Page 16 • Will radio waves travel through a hill that is in the way? Page 17 • How do I build a mesh network? Page 56 • What kind of antenna is the best one for my network? Page 102 • Can I build an access point using a recycled PC? Page 143 • How can I install Linux on my AP? Why should I do so? Page 152 Money • How can I know if a wireless link is achievable with a limited amount of money? Page 281 • Which is the best AP with the lowest price? Page 137 • How can I track and bill customers for using my wireless network? Page 165. 190 . Can I use a repeater in the middle to make it better? Page 77 • Should I use an amplifier instead? Page 115 Installation • How can I install my indoor AP on the top of a mast on my roof? Page 249 • Is it really useful to add a lightning protector and proper grounding to my antenna mast. spread all around the city? Page 53 • Is it true that I can reach a much greater distance adding a tin can or aluminum foil to my AP's antenna? Page 116 • Can I use wireless to connect to a remote site and share a single central Internet connection? Page 51 • My wireless link looks like it will be too long to work well. Is that true? Page 67 • How can I provide wireless connectivity to many remote clients.

See About This Book for more details. do I still need service from an ISP? Why? Page 27 • How many customers do I need to cover my costs? Page 287 • How many customers will my wireless network support? Page 65 • How do I make my wireless network go faster? Page 79 • Is my Internet connection as fast as it can be? Page 90 Security • How can I protect my wireless network from unauthorized access? Page 157 • Is it true that a wireless network is always insecure and open to attacks by hackers? Page 160 • Is it true that the use of open source software makes my network less secure? Page 167 • How can I see what is happening on my network? Page 174 Information and Licensing • What other books should I read to improve my wireless networking skills? Page 355 • Where can I find more information online? Page 349.net/ • Can I use parts of this book for my own teaching? Can I print and sell copies of this book? Yes.Chapter 1: Where to Begin 7 Partners and Customers • If I am supplying connectivity. http://wndw. .

.

a tree swaying in the wind. From a user s perspective. wireless connections are not particularly different from any other network connection: your web browser. But radio waves have some unexpected properties compared to Ethernet cable. But how do you know where the waves emanating from your wireless card are going? What happens when these waves bounce off of objects in the room or other buildings in an outdoor link? How can several wireless cards be used in the same area without interfering with each other? In order to build stable high-speed wireless links. and other applications all work as you would expect. the string of a guitar . it s very easy to see the path that an Ethernet cable takes: locate the plug sticking out of your computer. For example. it is important to understand how radio waves behave in the real world.2 A Practical Introduction to Radio Physics Wireless communications make use of electromagnetic waves to send signals across long distances. follow the cable to the other end.these are all examples of oscillations. 9 . What is a wave? We are all familiar with vibrations or oscillations in various forms: a pendulum. since the cables effectively keep their signals contained within the wire itself. and you ve found it! You can also be confident that running many Ethernet cables alongside each other won t cause problems. email.

you need to understand electromagnetic forces. then each wave will be twenty centimeters long: 1 meter/second = 5 cycles/second * W W = 1 / 5 meters W = 0.1. and this periodic change of air pressure then leaves the singers mouth and travels. In the case of electromagnetic waves. abbreviated Hz). A wave has a certain speed. for example from the top of one peak to the next. When such oscillations travel (that is.10 Chapter 2: A Practical Introduction to Radio Physics What they have in common is that something. if a wave on water travels at one meter per second. Simply drop a stone into the lake and you can see the waves as they move across the water over time. is swinging in a periodic manner. For example. some medium or object. at the speed of sound. ) is the distance measured from a point on one wave to the equivalent part of the next. wavelength. For example. and wavelength. The frequency is the number of whole waves that pass a fixed point in a period of time. frequency. Waves in water are easy to visualize. This kind of wave is sometimes called a mechanical wave. and it oscillates five times per second. the part that might be hardest to understand is: “What is it that is oscillating?” In order to understand that. . with a certain number of cycles per unit of time. which then travels across the lake as a wave. This is the distance from the center of the wave to the extreme of one of its peaks. The relationship between frequency. and can be thought of as the “height” of a water wave. These are connected by a simple relation: Speed = Frequency * Wavelength The wavelength (sometimes referred to as lambda. These oscillations periodically compress and decompress the air. frequency is measured in cycles per second (or Hertz. since it is defined by the motion of an object or its propagating medium. and amplitude are shown in Figure 2. and wavelength is measured in meters. Speed is measured in meters/second. A stone plunging into a lake causes a disturbance. a singer singing creates periodic oscillations in his or her vocal cords.2 meters = 20 cm Waves also have a property called amplitude. when the swinging does not stay bound to one place) then we speak of waves propagating in space.

This creates an electric field from plus to minus along the wire. For this wave. plus and minus). The magnetic force is the force between electrical currents. Electrons are particles that carry a negative electrical charge. or 2 Hz. and are radiated out into the space around the wire. the electrons have all been driven to the other side. but electrons are responsible for most of what we need to know about how radio behaves. periodically. the electric field vectors (arrows from plus to minus) are leaving the wire.Chapter 2: A Practical Introduction to Radio Physics time: 1 second wavelength ( ) 11 amplitude amplitude wavelength ( ) Figure 2. Let us look at what is happening in a piece of straight wire. Our most direct access to those is when our hand touches a door handle after walking on synthetic carpet. and the electric field points the other way. A more powerful example of electromagnetic forces is the lightning we see during thunderstorms. The next moment. There are other particles too. amplitude. At one moment. Electromagnetic forces Electromagnetic forces are the forces between electrical charges and currents. the top of the wire is negatively charged . Let us come back to the relation: Speed = Frequency * Wavelength . or brushing up against an electrical fence. so to speak.all the negative electrons are gathered there. What we have just described is known as a dipole (because of the two poles. in which we push the electrons from one and to the other and back. The electrical force is the force between electrical charges. or more commonly a dipole antenna. the frequency is 2 cycles per second. The motion of the electric field is commonly referred to as an electromagnetic wave. This is the simplest form of omnidirectional antenna.1: Wavelength. As this happens again and again. and frequency.

and engineering. math.000 km/s = 300.g.000. They are responsible for many of the differences between dif- . Let us take the example of the frequency of 802. Electromagnetic waves will even propagate through the vacuum of space.000 cycles / second wavelength lambda ( ) = = = = c / f 3*108 / 2.11b wireless networking. Powers of Ten NanoMicroMilliCentiKiloMegaGiga10-9 10-6 10-3 10-2 103 106 109 1/1000000000 1/1000000 1/1000 1/100 1 000 1 000 000 1 000 000 000 n μ m c k M G Knowing the speed of light. We will meet these terms again.4*109 1.4 GHz = 2. e. Powers of ten In physics. which is f = 2.000 m/s = 3*108 m/s c = f * Electromagnetic waves differ from mechanical waves in that they require no medium in which to propagate. from antennas that we build to objects that are in the way of the networks we intend to run. and so on. we often express numbers by powers of ten.25*10-1 m 12. Micro-seconds ( s). the speed is c.12 Chapter 2: A Practical Introduction to Radio Physics In the case of electromagnetic waves. the speed of light.000.5 cm Frequency and wavelength determine most of an electromagnetic wave s behavior. c = 300.400. we can calculate the wavelength for a given frequency. Centi-meters (cm). in Giga-Hertz (GHz).

direction of propagation electric field magnetic field Figure 2. If we put the antenna flat on the ground. you might have very little signal even though you have the strongest antennas. Therefore. Polarization Another important quality of electromagnetic waves is polarization. The electromagnetic spectrum Electromagnetic waves span a wide range of frequencies (and. This range of frequencies and wavelengths is called the electromagnetic spectrum.2: Electric field and complementary magnetic field components of an electromagnetic wave. We call this polarization mismatch. electrons only move up and down. not sideways (because there is no room to move) and thus electrical fields only ever point up or down.5*1014 Hz and 3. polarization becomes important when aligning antennas. The field leaving the wire and traveling as a wave has a strict linear (and in this case. the visible portion of the electromagnetic spectrum. wavelengths). Polarization describes the direction of the electrical field vector.8*1014 Hz. corresponding to wavelengths from circa 400 nm (violet/blue) to 800 nm (red). Polarization describes the orientation of the electric field. with the extremes of linear (only one direction) and circular polarizations (both directions at equal strength). we will always have some component of the field pointing other directions too.Chapter 2: A Practical Introduction to Radio Physics 13 ferent standards we might be choosing. Linear polarization is just one special case. If you imagine a vertically aligned dipole antenna (the straight piece of wire). If you ignore polarization. . we would find horizontal linear polarization. As one can imagine. an understanding of the basic ideas of frequency and wavelength helps a lot in practical wireless work. and is never quite so perfect: in general. vertically. Light lies roughly between the frequencies of 7. The part of the spectrum most familiar to humans is probably light. The most general case is elliptic polarization. accordingly. vertical) polarization.

5. Ultraviolet (on the higher frequencies side of visible light). In most countries. This region is called the ISM band. which uses a frequency around 100 MHz.with frequencies from about 1 GHz to 300 GHz.2. The frequencies most interesting to us are 2. which is used by the 802. Infrared (on the lower frequencies side of visible light).400 . but in the more narrow sense of the term. and Medical. Other commonly available equipment uses the 802. When talking about radio. the upper frequency limit would be 1 GHz.495 GHz. with license values being a huge economic factor. Most other parts of the electromagnetic spectrum are tightly controlled by licensing legislation. the ISM bands have been reserved for unlicensed use. These regions lie within the bands that are being kept open for general unlicensed use.14 Chapter 2: A Practical Introduction to Radio Physics We are also regularly exposed to other regions of the electromagnetic spectrum. This is true for the range from 3 Hz to 300 GHz. including Alternating Current (AC) or grid electricity at 50/60 Hz. This goes especially for those parts of the spectrum that are suitable for broadcast (TV. Radio is the term used for the portion of the electromagnetic spectrum in which waves can be generated by applying alternating current to an antenna.11a standard.11b and 802. which in fact works in exactly the same region as the wireless standards we are dealing with.150 . and many others. In between radio and infrared we find the region of microwaves . Scientific. which stands for Industrial. and wavelengths from 30 cm to 1 mm.11g radio standards (corresponding to wavelengths of about 12. The most popular use of microwaves might be the microwave oven. X-Rays / Roentgen radiation. Approximate frequency in Hz 104 106 108 1010 1012 1014 1016 1018 1020 1022 1024 microwave radio visible light infrared ultraviolet X rays gamma rays 104 102 100 10-2 10-4 10-6 10-8 10-10 10-12 10-14 10-16 Approximate wavelength in meters Figure 2.5 cm). radio) as well as voice and data communication. .3: The electromagnetic spectrum. many people think of FM radio. which operates at 5.850 GHz (corresponding to wavelengths of about 5 to 6 cm).

This is represented visually in Figure 2. as in “my Internet connection has 1 Mbps of bandwidth”. then the bandwidth would be 0.the more room in frequency space. The spectrum is divided into evenly sized pieces distributed over the band as individual channels. but are only separated by 5MHz. For a complete list of channels and their center frequencies for 802. meaning it can transmit data at 1 megabit per second.412 2 3 4 5 6 7 8 9 10 11 12 13 2.40 GHz to 2.4GHz band is used in 802.472 14 2.Chapter 2: A Practical Introduction to Radio Physics 15 Bandwidth A term you will meet often in radio physics is bandwidth.467 2. Note that channels 1.442 2.422 2.457 2. This means that adjacent channels overlap.447 2. the better it travels through and around things • The shorter the wavelength.462 2.08 GHz (or more commonly stated as 80MHz).11a. see Appendix B.427 2.417 2. If a range of 2. the more data you can fit in at a given moment.432 2. and can interfere with each other. 1 2. Note that channels are 22MHz wide. It is easy to see that the bandwidth we define here is closely related to the amount of data you can transmit within it . The term bandwidth is often used for something we should rather call a data rate.11b. Behavior of radio waves There are a few simple rules of thumb that can prove extremely useful when making first plans for a wireless network: • The longer the wavelength. the further it goes • The longer the wavelength. 6.4.452 2.4: Channels and center frequencies for 802. Bandwidth is simply a measure of frequency range.48 GHz is used by a device.437 2. and 11 do not overlap. Frequencies and channels Let us look a bit closer at how the 2.11b/g and 802.484 Channel Center Frequency (GHz) 22 MHz Figure 2. the more data it can transport .11b.

FM radio (88108MHz) can travel through buildings and other obstacles easily. For example.the place where you dip in . waves with longer wavelengths tend to travel further than waves with shorter wavelengths. a ship). wherever water particles are swinging and dancing. it would be well in the way of the wave. and which is extremely useful for understanding radio wave propagation. Dutch mathematician. The distance a wave can travel depends on the relationship between the wavelength of the wave and the size of obstacles in its path of propagation. Longer wavelength (and therefore lower frequency) waves tend to penetrate objects better than shorter wavelength (and therefore higher frequency) waves. while shorter waves (such as GSM phones operating at 900MHz or 1800MHz) have a harder time penetrating buildings. when comparing the range of an FM transmitter at 88MHz to the range at 108MHz. Longer waves pass around obstacles A wave on water which is 5 meters long will not be stopped by a 5 mm piece of wood sticking out of the water. a '0' or a '1'. causing the water to swing and dance. There is another principle that can be applied to all kinds of waves. named after Christiaan Huygens.in circles. This effect is partly due to the difference in power levels used for FM radio and GSM.g. This principle is known as the Huygens Principle. are rather easy to understand by example. It is harder to visualize waves moving “through” solid objects. a 'yes' or a 'no'. Shorter waves can carry more data The faster the wave swings or beats.16 Chapter 2: A Practical Introduction to Radio Physics All of these rules. but is also partly due to the shorter wavelength of GSM signals. they will cause their neighbor particles to do . but this is the case with electromagnetic waves. Waves will leave the center of the stick . Lower frequency transmitters tend to reach much greater distances than high frequency transmitters at the same power. the more information it can carry every beat or cycle could for example be used to transport a digital bit. If instead the piece of wood were 50 meters big (e.1695. Now. Longer waves travel further Assuming equal power levels. simplified as they may be. This effect is often seen in FM radio. physicist and astronomer 1629 . Imagine you are taking a little stick and dipping it vertically into a still lake's surface.

such as diffraction. Actually. • Water. the need for line of sight as well as the fact that sometimes we seem to be able to go around corners. we may well consider metal and water perfect absorbers: we will not be able to go through them (al1.” This principle holds true for radio waves as well as waves on water. with no line of sight. the Huygens principle. Absorption When electromagnetic waves go through 'something' (some material). Electrons can move freely in metals. thus taking away some of the wave s energy1. A commonly held myth is that water “resonates” at 2. 2. water doesn t appear to have any particular “resonant” frequency. Let us now look into what happens to electromagnetic waves as they travel. Microwaves cause water molecules to jostle around. Clear window glass is obviously transparent for light.4 GHz is an unlicensed ISM frequency.4 GHz. a new circular wave will start. in simple form. How much they lose in power will depend on their frequency and of course the material. and are readily able to swing and thus absorb the energy of a passing wave.only for light the wavelength is far too short for human beings to actually see the effects directly. for sound as well as light . It recognizes that each point of an advancing wave front is in fact the center of a fresh disturbance and the source of a new train of waves.Chapter 2: A Practical Introduction to Radio Physics 17 the same: from every point of disturbance. and will heat when in the presence of high power radio waves at just about any frequency. For the purpose of practical wireless networking. This is. which is why that frequency is used in microwave ovens. Often. and so was a good political choice for use in microwave ovens. an absorption coefficient is used to describe a material s impact on radiation. and that the advancing wave as a whole may be regarded as the sum of all the secondary waves arising from points in the medium already traversed. In the words of wikipedia. while the glass used in sunglasses filter out quite a share of the light intensity and also the ultraviolet radiation. For microwaves. This principle will help us to understand diffraction as well as Fresnel zones. Water spins and jostles around near radio. the two main absorbent materials are: • Metal. . they generally get weakened or dampened. This view of wave propagation helps better understand a variety of wave phenomena.org: “The Huygens' principle is a method of analysis applied to problems of wave propagation in the far field limit.

They have a strong influence. One simple method of measuring the absorption of plastic at 2. as long as the distance between bars is small compared to the wavelength. They are to microwave what a brick wall is to light. For trees and wood. and therefore at different times) play such an important role in wireless networking. radio waves are reflected when they come in contact with materials that are suited for that: for radio waves. it is always a good idea to measure and verify that the material does not absorb radio energy around 2. Note that in the eyes of a radio wave. Before you build a component from plastic (e. There are other materials that have a more complex effect on radio absorption. with waves and ripples changing all the . Although the rules of reflection are quite simple. the main sources of reflection are metal and water surfaces. This explains why multipath effects (i. As far as radio networking is concerned. signal reaching their target along different paths. a one cm metal grid will act much the same as a metal plate. At 2. libraries. wet fresh wood will absorb a lot. Plastics and similar materials generally do not absorb a lot of radio energybut this varies depending on the frequency and type of material. The rules for reflection are quite simple: the angle at which a wave hits a surface is the same angle at which it gets deflected. and in many circumstances a change in weather can bring a radio link down. The same goes for hotspots. and outdoor installations. weather protection for a radio device and its antennas).18 Chapter 2: A Practical Introduction to Radio Physics though thin layers of water will let some power pass). low clouds and so forth all will be in the way of radio links.4 GHz is to put a sample in a microwave oven for a couple of minutes. cafe installations.g. If the plastic heats up. Reflection Just like visible light. When talking about water.4 GHz. Old dead dry wood is more or less transparent. let us talk about ourselves: humans (as well as other animals) are largely made out of water. Lastly. Water surfaces.e. we have to remember that it comes in different forms: rain. the amount of absorption depends on how much water they contain. we may well be described as big bags of water. The same goes for urban situations: look around you in city environment and try to spot all of the metal objects.4 GHz. fog and mist. things can become very complicated when you imagine an office interior with many many small metal objects of various complicated shapes. Orienting an office access point in such a way that its signal must pass through many people is a key mistake when building office networks. a dense grid of bars acts just the same as a solid surface. with the same strong absorption. then it absorbs radio energy and should not be used for weatherproofing.

like a small door. just like a wave that we see rolling onto an ocean beach. in its way to block it.g. it would be hard to explain how it can reach points that should be hidden by a barrier. the phenomenon makes sense. A parabolic uses this effect to concentrate radio waves spread out over its surface in a common direction.and it might just as well be an electromagnetic wave . a circular wave will start.Chapter 2: A Practical Introduction to Radio Physics 19 time. . i r i= r Figure 2. We should also add that polarization has an impact: waves of different polarization in general will be reflected differently. If you look at this wavefront . Imagine a wave on water traveling in a straight wave front. effectively make for a very complicated reflection object which is more or less impossible to calculate and predict precisely. and it will of course reach points that are not in a direct line behind this opening. but also on either side of it. When modeled as a wavefront.as a beam (a straight line). It is the effect of “waves going around corners”. say a wooden solid fence. Now we put a solid barrier. We use reflection to our advantage in antenna building: e. From this opening.5: Reflection of radio waves. Diffraction Diffraction is the apparent bending of waves when hitting an object. We cut a narrow slit opening into that wall. The angle of incidence is always equal to the angle of reflection. we put huge parabolas behind our radio transmitter/receiver to collect and bundle the radio signal into a fine point.

waves will “bend” around corners or through an opening in a barrier. every point on a wavefront can be considered the starting point for a spherical “wavelet”.7: The Huygens Principle. Through means of the effect of diffraction. But for our purposes. . mountain peaks.6: Diffraction through a narrow slit. This idea was later extended by Fresnel. with a wavelength of several centimeters. Imagine that at any given instant. will show the effects of diffraction when waves hit walls. Microwaves. It seems as if the obstruction causes the wave to change its direction and go around corners.20 Chapter 2: A Practical Introduction to Radio Physics Diffraction Straight wave front Figure 2. The wavelengths of visible light are far too small for humans to observe this effect directly. and whether it adequately describes the phenomenon is still a matter of debate. Diffraction Potential spherical wavelets Figure 2. and other obstacles. the Huygens model describes the effect quite well. The Huygens Principle provides one model for understanding this behavior.

Chapter 2: A Practical Introduction to Radio Physics 21 Figure 2. When peak hits peak. Interference When working with waves.8: Diffraction over a mountain top.you will see that where two waves cross. It can also result in zero. But in some very specific applications.destructive interference. Note that diffraction comes at the cost of power: the energy of the diffracted wave is significantly less than that of the wavefront that caused it. you will have complete annihilation ((1 + (-)1 = 0) . you can take advantage of the diffraction effect to circumvent obstacles.9: Constructive and destructive interference. + = + = Figure 2. This is called constructive interference. This is easy to understand when you draw two sine waves and add up the amplitudes. one plus one does not necessarily equal two. . you will have maximum results (1 + 1 = 2). there will be areas of higher wave peaks and others that remain almost flat and calm. When peak hits valley. You can actually try this with waves on water and two little sticks to create circular waves .

their beams are a lot wider . and other sources of microwave. the wave is annihilated and no signal can be received. This is also the case for the widening of waves as they travel. Whenever waves of equal amplitude and opposite phase cross paths. is quite easy to understand when talking about visible light: if we can see a point B from point A where we are. Rather than pointing at the moon. You can see this effect for yourself using an inexpensive laser pointer and a pair of binoculars on a clear night. the word Interference is typically used in a wider sense. neighboring channels. so to speak.g. often abbreviated as LOS. Consequently. In wireless technology. and if you let them travel long enough. The much more common case is that waves will combine to form a completely garbled waveform that cannot be effectively used for communication. So. Light has a wavelength of about 0. you can see the results despite of their short wavelength. this means fixed positions from the peaks of the one wave to the other's.22 Chapter 2: A Practical Introduction to Radio Physics In order for whole trains of waves to add up or cancel each other out perfectly. but does not completely eliminate it. Remember that most propagation characteristics of electromagnetic waves scale with their wavelength. especially in urban environments or closed spaces (such as a conference space) where many networks may compete for use of the spectrum. Things get a bit more complicated when we are dealing with microwaves.they need more space. e. they would have to have the exact same wavelength and a fixed phase relation. microwaves as used in wireless networking have a wavelength of a few centimeters. Line of sight The term line of sight. The radius of your beam will increase as the distance increases.5 micrometers. we have line of sight. Simply draw a line from A to B. point at a distant mountain or unoccupied structure (such as a water tower). . its beam will widen to well over 100 meters in radius by the time it reaches the surface. and if nothing is in the way. The modulation techniques and use of multiple channels help to deal with the problem of interference. for disturbance through other RF sources. we have line of sight. When pointing a well focussed laser at the moon. when wireless networkers talk about interference they typically talk about all kinds of disturbance by other networks. Interference is one of the main sources of difficulty in building wireless links. Note that visible light beams widen just the same.

Understanding the Fresnel zone The exact theory of Fresnel (pronounced “Fray-nell”) zones is quite complicated. Note that this gives you the radius . so usually in wireless networking we check that about 60 percent of the radius of the first Fresnel zone should be kept free. e. their path is longer. but we are chiefly concerned with zone 1. Taking this approach and calculating accordingly. Fresnel zone theory simply looks at a line from A to B. introducing a phase shift between the direct and indirect beam. a tree or a building. d is the total link distance in meters. We know that waves of one frequency can interfere with each other.where r is the radius of the zone in meters.10: The Fresnel zone is partially blocked on this link. Of course... However. and f is the frequency in MHz. the concept is quite easy to understand: we know from the Huygens principle that at each point of a wavefront new circular waves start. If this area were partially blocked by an obstruction.31 * sqrt((d1*d2)/(f*d)) .Chapter 2: A Practical Introduction to Radio Physics 23 The line of sight that we need in order to have an optimal wireless connection from A to B is more than just a thin line . you find there are ring zones around the direct line A to B which contribute to the signal arriving at point B. although the visual line of sight appears clear. Fresnel radius Line of sight Partial obstruction Figure 2. nothing is ever perfect. Whenever the phase shift is one full wavelength. we therefore need to be sure that these zones be kept free of obstructions. Consequently.its shape is more like that of a cigar. and then at the space around that line that contributes to what is arriving at point B. d1 and d2 are distances from the obstacle to the link end points in meters. Here is one formula for calculating the first Fresnel zone: r = 17. an ellipse. Note that there are many possible Fresnel zones. When building wireless links.g. Its width can be described by the concept of Fresnel zones. you get constructive interference: the signals add up optimally. Some waves travel directly from A to B. the signal arriving at the far end would be diminished. We know that microwave beams widen as they leave the antenna. while others travel on paths off axis.

you need to subtract the result from a line drawn directly between the tops of the two towers.16 meters above ground level in the middle of the link. But how tall could a structure at that point be to clear 60% of the first zone? r = 0.11b channel 6): r = 17.437 GHz (802. The electric field is measured in V/m (potential difference per meter). let s calculate the size of the first Fresnel zone in the middle of a 2km link.3 meters tall at the center of the link would block up to 40% of the first Fresnel zone.31 sqrt((1000 * 1000) / (2437 * 2000)) r = 17. an antenna and a voltmeter.g. gains and radio sensitivity in Chapter 3. the power contained within it is proportional to the square of the electric field P ~ E2 Practically. the first Fresnel zone would pass just 2.31 sqrt(1000000 / 4874000) r = 7.6 * 17. we measure the power by means of some form of receiver. we can see that a structure 5. oscilloscope. Looking at the signal s power directly means looking at the square of the signal in Volts.24 Chapter 2: A Practical Introduction to Radio Physics of the zone.70 meters Subtracting the result from 10 meters. or even a radio card and laptop. Power Any electromagnetic wave carries energy . To calculate the height above ground. This is normally acceptable. losses.6 * 17. The power P is of key importance for making wireless links work: you need a certain minimum power in order for a receiver to make sense of the signal. We will come back to details of transmission power. .31 sqrt(600000 / 4874000) r = 4. not the height above ground. but to improve the situation we would need to position our antennas higher up.we can feel that when we enjoy (or suffer from) the warmth of the sun. e. For example. transmitting at 2.31 sqrt((1000 * 1000) / (2437 * 2000)) r = 0. Here we will briefly discuss how the power P is defined and measured. power meter. The amount of energy received in a certain amount of time is called power. or change the direction of the link to avoid the obstacle.84 meters Assuming both of our towers were ten meters tall.

it is just a convenient method which makes calculations a lot simpler. There is no new physics hidden in this . The result would be: 0 1 2 3 4 n meters meter meters meters meters meters = = = = = = 1 (full signal) 1/2 1/4 1/8 1/16 1/2n = 2-n This is exponential behavior. It is defined by: dB = 10 * Log (P1 / P0) where P1 and P0 can be whatever two values you want to compare. Another example of a dimensionless unit is the percent (%) which can also be used in all kinds of quantities or numbers. we just add. is absorption. it defines a relationship between two measurements of power. Why are decibels so handy to use? Many phenomena in nature happen to behave in a way we call exponential. For example. dimensionless units represent a relationship. and each meter of wall takes away half of the available signal. While measurements like feet and grams are fixed. things become a lot easier: instead of taking a value to the n-th power. The decibel is a dimensionless unit2. this will be some amount of power. in our case. . Another example. Typically. Suppose a wall is in the path of our wireless link. we just multiply by n. the human ear senses a sound to be twice as loud as another one if it has ten times the physical signal. quite close to our field of interest. Here are some commonly used values that are important to remember: +3 -3 +10 -10 dB dB dB dB = = = = double power half the power order of magnitude (10 times power) one tenth power 2. But once we have used the trick of applying the logarithm (log). that is.Chapter 2: A Practical Introduction to Radio Physics 25 Calculating with dB By far the most important technique when calculating power is calculating with decibels (dB). Instead of multiplying values.

predictable path. there are a number of relative definitions that are based on a certain base value P0. . It is approximated by a dipole. By now you should understand that radio waves don t travel in a straight. Here are equivalent power levels expressed in milliwatts and dBm: 1 2 100 1 mW mW mW W = = = = 0 dBm 3 dBm 20 dBm 30 dBm Physics in the real world Don t worry if the concepts in this chapter seem challenging. Most people find it difficult to understand phenomenon that they can t even see with their own eyes. Another common (although less convenient) convention for expressing power is in milliwatts. and predict how the waves will travel along the way. see the resources list in Appendix A. Understanding how radio waves propagate and interact with the environment is a complex field of study in itself. For more information about this evolving field. To make reliable communication networks.26 Chapter 2: A Practical Introduction to Radio Physics In addition to dimensionless dB. The isotropic model is useful for describing the relative power gain of a real world antenna. but a perfect isotropic antenna cannot be built in reality. you will need to be able to calculate how much power is needed to cross a given distance. The most relevant ones for us are: dBm dBi relative to P0 = 1 mW relative to an ideal isotropic antenna An isotropic antenna is a hypothetical antenna that evenly distributes power in all directions. There is much more to learn about radio physics than we have room for here.

we will begin with a review of the networking concepts that define TCP/IP. The network design you choose to implement should fit the communications problem you are trying to solve. Finally. By understanding TCP/IP. routing. the primary family of networking protocols currently used on the Internet.3 Network Design Before purchasing equipment or deciding on a hardware platform. Most likely. Do you need to connect a remote site to an Internet connection in the center of your campus? Will your network likely grow to include several remote sites? Will most of your network components be installed in fixed locations. We will then see examples of how other people have built wireless networks to solve their communication problems. you should have a clear idea of the nature of your communications problem. you can build networks that will scale to virtually any size. If you are already comfortable with the essentials of TCP/IP networking (including addressing. firewalls. and routers). switches. or will your network expand to include hundreds of roaming laptops and other devices? In this chapter. and will ultimately become part of the global Internet. we will present several common methods for getting your information to flow efficiently through your network and on to the rest of the world. you are reading this book because you need to connect computer networks together in order to share resources and ultimately reach the larger global Internet. including diagrams of the essential network structure. Networking 101 TCP/IP refers to the suite of protocols that allow conversations to happen on the global Internet. you may want 27 .

customs. USA. USA. After several days. Italy is a fantastic city to get lost in. Once it clears through U. The artist first packs the mask into a shipping box and addresses it to the office in Seattle. A clerk at the office . who attaches some official forms and sends it to a central package processing hub for international destinations.1: Another kind of network mask. Imagine a tourist who happens to find papier-mâché mask as a souvenir. The package eventually makes its way onto a delivery van which has a route that brings it to the proper address. specializing in delivery to only one or two of the six sestieri (districts) of Venice. Figure 3. We will now review the basics of Internet networking. arriving at a central import processing location in the U.S. This may sound like an ordinary (or even trivial) task. then on to the Seattle postal processing center.S. but let's look at what actually happens.. Postal carriers in Venice are some of the most highly trained in the world. Introduction Venice. Many people find that knowing the location of the water and the sun is far more useful than trying to find a street name on a map. in the proper neighborhood. and never go in a simple straight line.S. This is necessary due to the intricate layout of that ancient city. The roads are mere foot paths that cross water in hundreds of places. Polo.28 Chapter 3: Network Design to skip ahead to Designing the Physical Network on Page 51. on the proper street. the package clears Italian customs and finds its way onto a transatlantic flight. Venezia to an office in Seattle. They then hand this off to a postal employee. and wants to have it shipped from the studio in S. the package is sent to the regional distribution point for the northwest U.

His job is to pick up packages from his local neighborhood and forward them to the next closest hub in the delivery chain. Packets are forwarded between routers until they reach their ultimate destination. the best route to a few other local networks. how to get to the local network.jpg Part 1 of 10 Computer Image. This list of possible routes is called the routing table. the destination address is examined and compared against its internal routing table. The router needs only to keep track of a handful of routes (for example.2: Internet networking.jpg Part 10 of 10 Server Figure 3. and deliver them to the proper person. and are labeled with their source and destination. Packages can only make their way through the international postal system because we have established a standardized addressing scheme for packages. the destination address must be written legibly on the front of the package. and include all critical information (such as the recipient's name. His job is simply to accept packages as they arrive.Chapter 3: Network Design 29 accepts the package and puts it in the proper incoming mail box. the postal carrier in Venice has no need to worry about how to get to the correct neighborhood in Seattle. which is often its own Internet gateway (via the default route). The clerk at the office in Seattle neither knows nor cares about how to get to the sestiere of S. until the packet eventually arrives at its destination. For example. If the router has no explicit route to the destination in question. it sends the packet to the closest match it can find. Similarly. This is very similar to how Internet routing works. Venezia. As packets arrive at the router. Internet Router Router Image. Polo. Once it arrives. The computer then sends these packets to a router. and one route to a gateway to the rest of the Internet). And the next router does the same. and so forth. the package is retrieved and the mask itself is finally received. which decides where to send them next. . A message is split up into many individual packets.

iso. Without this information. it leaves the actual implementation details up to the manufacturer. But once the communication becomes more complex than a simple conversation between two people.30 Chapter 3: Network Design street address. Each layer can be implemented in hardware (more common for lower layers) or software. but simply delegates each communications "job" to a single layer within a welldefined hierarchy. Each layer is independent of the layers around it. The full standard is available as publication "ISO/IEC 7498-1:1994. The OSI model The international standard for Open Systems Interconnection (OSI) is defined by the document ISO/IEC 7498-1. country.org/ittf/PubliclyAvailableStandards/. packages are either returned to the sender or are lost in the system. Packets can only flow through the global Internet because we have agreed on a common addressing scheme and protocol for forwarding packets. The most well-known of these is the OSI model. The abstraction between layers makes it easy to design elaborate and highly reliable protocol stacks. and each builds on the services provided by the layer below while providing new services to the layer above. such as the ubiquitous TCP/IP stack. Without a common set of communication protocols to regulate when and how each computer can speak. While the ISO/IEC 7498-1 specification details how layers should interact with each other." available from http://standards. protocol becomes just as important as language. but without a set of rules in place to establish who has the right to use the microphone. A protocol stack is an actual implementation of a layered communications framework. the communication of an individual s ideas to the entire room is nearly impossible. As long as the interface between layers adheres to . and postal code). the Internet would be a chaotic mess where every machine tries to speak at once. as outlined by the International Standards Organization and the International Electrotechnical Commission. All of the people in an auditorium may speak English. Cooperative communications Communication is only possible when the participants speak a common language. Now imagine an auditorium as big as the world. The OSI model doesn't define the protocols to be used in a particular network. These standard communication protocols make it possible to exchange information on a global scale. full of all of the computers that exist. The OSI model divides network traffic into a number of layers. People have developed a number of communications frameworks to address this problem. city.

Examples of protocols that operate at this layer are TCP and UDP. 6 Presentation The Presentation Layer deals with data representation. The human sits above this layer. Some protocols at the transport layer (such as TCP) ensure that all of the data has arrived at the destination. NetBIOS and RPC are two examples of a layer five protocol. IP (the Internet Protocol) is the most common Network Layer protocol. formatting checks. Packets can leave the link local network and be retransmitted on other networks. Another critical Network Layer protocol is ICMP. data compression. The Transport Layer provides a method of reaching a particular service on a given network node. interacting with the application. This layer is also sometimes referred to as the Internet Layer. implementers are free to use whatever means are available to build their protocol stack. The Session Layer manages the logical communications session between applications. and is the level at which human communication happens. HTTP. one on each of the networks to be interconnected. 5 Session 4 Transport 3 Network . Here is a brief outline of the seven-layer OSI networking model: Layer 7 Name Application Description The Application Layer is the layer that most network users are exposed to. etc. UDP is a "connectionless" protocol commonly used for video and audio streaming. which is a special protocol which provides various management messages needed for correct operation of IP. and SMTP are all application layer protocols. This is the layer where routing occurs. Nodes on the Internet are reached by their globally unique IP address.Chapter 3: Network Design 31 the standard. This means that any given layer from manufacturer A can operate with the same layer from manufacturer B (assuming the relevant specifications are implemented and interpreted correctly). and is reassembled and delivered to the next layer in the proper order. This would include MIME encoding. before it reaches the application. byte ordering. FTP. Routers perform this function on a network by having at least two network interfaces.

or a room full of wireless devices all using the same radio channel) they use the Data Link Layer to communicate. so it must implement all seven layers. This layer is sometimes known as the Media Access Control (MAC) layer. Common examples of data link protocols are Ethernet. with the foundation at layer one. The first three layers (Physical. A web server or a laptop computer runs applications. with seven at the top. This is meant to reinforce the idea that each layer builds upon. A network switch can only distribute packets by using MAC addresses. or just about any other medium capable of transmitting signals. radio waves. Cut wires. activity at these layers is determined by the configuration of cables. if the fourth floor is on fire. since all nodes connected at this layer communicate with each other directly." That is. such as the name of a website. the layers below. This can be a copper CAT5 cable. so it need implement only layers one through three. The Physical Layer is the lowest layer in the OSI model. If you remove any single layer. and similar devices. several computers plugged into a hub. Communication on this layer is said to be link-local. and is widely regarded as the complete and definitive network model. so it need only implement layers one and two. the building will not stand. It provides a framework for manufac- . broken fiber. This is a unique 48 bit number assigned to every networking device when it is manufactured. Data Link. and the roof at layer seven. and the wireless networking protocols (802. On networks modeled after Ethernet. The OSI model is internationally recognized.11a/b/g). Some advanced routers may implement layer four and above. to allow them to make decisions based on the higher-level information content in a packet. 1 Physical The layers in this model are numbered one through seven. a fiber optic bundle. and refers to the actual physical medium over which communications take place.32 Chapter 3: Network Design Layer 2 Name Data Link Description Whenever two or more nodes share the same physical medium (for example. switches. Similarly. ATM. Token Ring. and RF interference are all physical layer problems. and Network) all happen "on the network. or the attachments of an email. then nobody can pass through it in either direction. the next layers as successive floors. nodes are referred to by their MAC address. routers. and depends upon. Imagine the OSI model as a building. A simple router can route packets using only their IP addresses.

it is often used as a pragmatic model for understanding and troubleshooting Internet networks. The vast majority of the Internet uses TCP/IP. The person first needs to interact with the road itself (the Physical layer). In particular. go to the proper floor and room num- . turn at the proper place to connect to other roads and arrive at the correct address (the Internet layer). and everything above layer four tends to apply to specific applications. The TCP/IP model of networking describes the following five layers: Layer 5 4 3 2 1 Name Application Transport Internet Data Link Physical In terms of the OSI model. pay attention to other traffic on the road (the Data Link layer). Since the first three layers are interoperable between virtually all manufacturers' equipment. the OSI model can be simplified into a smaller collection of five layers. We will use the TCP/IP model when discussing networks in this book. Many network engineers think of everything above layer four as "just data" that varies from application to application. layers five through seven are rolled into the topmost layer (the Application layer).Chapter 3: Network Design 33 turers and network protocol implementers that can be used to build networking devices that interoperate in just about any part of the world. the OSI model can seem needlessly complex. For the majority of Internet network implementations. The first four layers in both models are identical. people who build and troubleshoot TCP/IP networks rarely need to deal with problems at the Session or Presentation layers. The TCP/IP model can be compared to a person delivering a letter to a downtown office building. and layer four works between all hosts using TCP/IP. the TCP/IP model is not an international standard and its definitions vary. this simplified model works well when building and troubleshooting TCP/IP networks. From the perspective of a network engineer or troubleshooter. Nevertheless. and so we can make some assumptions about networks that make them easier to understand. The TCP/IP model Unlike the OSI model.

The acronym stands for Transmission Control Protocol (TCP) and Internet Protocol (IP). but actually refers to a whole family of related communications protocols.250. and it operates at layers three and four of the TCP/IP model. The authority assigns large blocks of consecutive addresses to smaller authorities. In this discussion. 192. the delivery person is free to go on their way. IP Addressing In an IPv4 network. A group of related addresses is referred to as an address space. or subnets for short. Each of the usable IP addresses is a unique identifier that distinguishes one network node from another. IP addresses must be unique and generally cannot be used in different places on the Internet at the same time. TCP/IP is also called the Internet protocol suite. they would range from 0.255. we will focus on version four of the IP protocol (IPv4) as this is now the most widely deployed protocol on the Internet.23. otherwise.34 Chapter 3: Network Design ber (the Transport layer).1.0. or to their customers. Examples of IP addresses are 10. and finally give it to a receptionist who can take the letter from there (the Application layer). the address is a 32-bit number.1. Once they have delivered the message to the receptionist. . routers would not know how best to route packets to them. IP addresses are allocated by a central numbering authority that provides a consistent and coherent numbering method.168.” which of course stands for “Physical / Data Link / Internet / Transport / Application. who in turn assign smaller consecutive blocks within these ranges to other authorities.625). although many of these are reserved for special purposes and should not be assigned to hosts.228.16. This ensures that duplicate addresses are not used by different networks. Interconnected networks must agree on an IP addressing plan.0.” The Internet protocols TCP/IP is the protocol stack most commonly used on the global Internet. If you enumerated every possible IP address. The five layers can be easily remembered by using the mnemonic “Please Don t Look In The Attic. These groups of addresses are called sub-networks. Large subnets can be further subdivided into smaller subnets.17.0 to 255.5.1.0. normally written as four 8-bit numbers expressed in decimal form and separated by periods. or 172.255. This yields a total of more than four billion possible IP addresses (255 x 255 x 255 x 255 = 4.255.

2 PC Server 10. A subnet mask determines the size of a given network.1.1. you can logically define both a host and the network to which it belongs. Not all RFCs are actual standards. However. the first value is taken as the network address (.3: Without unique IP addresses. which server will it reach? Subnets By applying a subnet mask (also called a network mask. By convention. In binary notation. This leaves 254 addresses available for hosts on this network.1.1. 255. 255. etc.255. RFC is short for Request For Comments.net/ . You will find this notation used when configuring network interfaces.255 or 11111111). much like an IP address. Using a /24 netmask.0 or 00000000). creating routes. 8 bits are reserved for hosts (32 bits total .0 can be simplified as /24. and is defined in RFC15181 .0 is one common netmask. The result is "1" if both of the bits being compared are 1. A logical AND is performed by comparing two bits.255. subnet masks are more succinctly expressed using CIDR notation. For example. which simply enumerates the number of bits in the mask after a forward slash (/). unambiguous global routing is impossible. Thus. the "1" bits in the mask indicate the network address portion. CIDR is short for Classless Inter-Domain Routing.255. Subnet masks work by applying AND logic to the 32 bit IP number.2. This yields up to 256 possible host addresses (28 = 256).255. subnet masks are expressed using dotted decimal form. and "0" bits indicate the host address portion. Traditionally.2 Figure 3. RFCs can be viewed online at http://rfc. or simply netmask) to an IP address. and the last value is taken as the broadcast address (.1.1.24 bits of netmask = 8 bits for hosts). If the PC requests a web page from 10.Chapter 3: Network Design 35 ? Internet Server 10. RFCs are a numbered series of documents published by the Internet Society that document ideas and concepts related to Internet technologies.

00001010.255.10: 00001010.0: 00001010.00001010.255.255.0.10.0/24.0 (or /15 CIDR).10. common netmasks include: .0 as the network address and 10.1.255.00001010.10. first convert everything to binary.10.0 in binary contains twenty-four "1" bits: 255 255 255 0 11111111.10.255. 10.254. This is a large block. and each is a valid subnet whose size is a power of two.0-10.00001010.255. This network consists of the hosts 10.00000000 This results in the network 10.0.1.10. It could even be subdivided into a mixture of different block sizes. for example into 512 subnets of 256 addresses each.10. Alternatively. While many netmasks are possible. Here are all of the possible outcomes of a binary AND comparison between two bits.255.0.255. The netmask 255.00001010 255.36 Chapter 3: Network Design also "1".0: 11111111.10.072 addresses. then 10.536 addresses.0.1. and so on up to 10. Subnet masks are not limited to entire octets.0 to 10.10.10.1.0. or in many other ways.10.0.255.0.10.11111111. containing 131.0.254. Otherwise the result is "0".0.1. or 8192 blocks of 16 addresses. The first one would be 10. we can apply a logical AND to each of the bits to determine the network address.255 as the broadcast address.00000000 When this netmask is combined with the IP address 10.10. as long as none of them overlap.11111111. One can also specify subnet masks like 255.10. It could be further subdivided.255.10.0-10.11111111.255. with 10.10.11111111.0-10.00000000 ----------------------------------10. it could be subdivided into 2 blocks of 65.1 through 10.10. from 10. Bit 1 0 0 1 1 Bit 2 0 1 0 1 Result 0 0 0 1 To understand how a netmask is applied to an IP address.

255.org/).0. There are three common netmasks that have special names. usually /8 subnets with 16 million addresses each.192 255.255.0.224 255.0.255. .255. The ISP then allocates smaller IP blocks to their clients as required.255. The 4 billion available IP addresses are administered by the Internet Assigned Numbers Authority (IANA.255. and a /24 (255.0 # of Hosts 4 8 16 32 64 128 256 65 536 16 777 216 With each reduction in the CIDR value the IP space is doubled. A /16 (255. Global IP Addresses Have you ever wondered who controls the allocation of IP space? Globally routable IP addresses are assigned and distributed by Regional Internet Registrars (RIRs) to ISPs. but are still often used for historical reasons.0 255.0) defines a Class A network.0 255.255. These names were around long before CIDR notation.0.255. IANA has divided this space into large subnets.255.0.Chapter 3: Network Design 37 CIDR /30 /29 /28 /27 /26 /25 /24 /16 /8 Decimal 255.128 255.252 255.0) is called a Class C.255.255.255.0) is a Class B. These subnets are delegated to one of the five regional Internet registries (RIRs).255. http://www. Virtually all Internet users obtain their IP addresses from an ISP.iana. which are given authority over large geographic areas.248 255.0.255.255. Remember that two IP addresses within each network are always reserved for the network and broadcast addresses.255.240 255.255. A /8 network (with a netmask of 255.255.

The registry system assures that IP addresses are not reused in any part of the network anywhere in the world.net/) Your ISP will assign globally routable IP address space to you from the pool allocated to it by your RIR.).arin. . and typically serve information to other machines (such as email services. The process of moving packets between networks is called routing. http://www. The five RIRs are: • African Network Information Centre (AfriNIC.apnic. http://www. Once IP address assignments have been agreed upon. http://www. etc. it is possible to pass packets between networks and participate in the global Internet. Static IP addresses are important because servers using these addresses may have DNS mappings pointed towards them. Static IP Addresses A static IP address is an address assignment that never changes.net/) • Réseaux IP Européens (RIPE NCC.net/) • American Registry for Internet Numbers (ARIN.afrinic.net/) • Asia Pacific Network Information Centre (APNIC.lacnic. http://www.net/) • Regional Latin-American and Caribbean IP Address Registry (LACNIC.4: Authority for Internet IP address assignments is delegated to the five Regional Internet Registrars. web servers.ripe.38 Chapter 3: Network Design RIPE ARIN LACNIC AfriNIC APNIC Figure 3. http://www.

Private IP addresses Most private networks do not require the allocation of globally routable. In particular. Assigning addresses dynamically allows ISPs to save money. the European RIR) are very strict on IP address usage for ISP's. either by request or automatically depending on your means of connection to the Internet. and 192. These addresses are not intended to be routed on the Internet. Organizations typically use IP addresses from the private address space for machines on the internal network. such as a home computer which is on a dial-up connection.0. They only need an address for each customer who is active at any one time. Globally routable IP addresses cost money. the node may receive the same IP address or a different one from the pool of available addresses. Dynamic IP Addresses Dynamic IP addresses are assigned by an ISP for non-permanent nodes connecting to the Internet. and they will often charge extra to provide a static IP address to their customers. Dynamic IP addresses can be assigned automatically using the Dynamic Host Configuration Protocol (DHCP).168. because it enables them to use fewer IP addresses than their total number of customers. These are defined in RFC1918.0. Upon renewal. depending on the type of Internet connection.0/16. IP addresses assigned by DHCP are valid for a specified time (called the lease time). IP addresses can be assigned randomly from a pool by your ISP.16. There are currently three blocks of private address space reserved by IANA: 10. 172. or might be assigned according to a policy. A node using DHCP first requests an IP address assignment from the network.0/8. and are typically unique only within an organization or group of organizations which choose to follow the same numbering scheme. .0. or the Point-to-Point Protocol (PPP). Dynamic addresses are popular with Internet service providers. and automatically configures its network interface. The node must renew the DHCP lease before the lease time expires. public IP addresses for every computer in the organization.0.Chapter 3: Network Design 39 Blocks of static IP addresses may be assigned by your ISP. computers which are not public servers do not need to be addressable from the public Internet.0/12. and some authorities that specialize in the assignment of addresses (such as RIPE.

). etc. Some Internet providers may allocate private addresses like these instead of public addresses to their customers. They use the corresponding IP addresses 192. If you ever intend to link together private networks that use RFC1918 address space.1.0/24 To LAN To LAN Figure 3. 10.) or into blocks of any other logical size.3. their private addresses must be translated to public addresses. and is normally performed at the gateway between the private network and the Internet.2.1. B. These hosts are part of a /24 network (their network mask is 255. 10. The network administrators at each location can then break the network down further into multiple Class C networks (10. then all of the machines will be reachable from any point in the network without having to renumber network devices. dormitories. or VPN).1.2. wireless link.1.1.168.0.0/16 Router 192.0. be sure to use unique addresses throughout all of the networks.16.5: RFC1918 private addresses may be used within an organization. although this has serious disadvantages.1.2 and 192.255.0/24 Router 10. . 192.168.6.1.1.168.15. field office two. and so forth).99. and are not routed on the global Internet.1. Routing Imagine a network with three hosts: A.255.1.0). and are not directly reachable from it.0/24. and C. One block could be assigned to each network according to its physical location (the campus main branch.2. For example.0/24 Router Internet Router 172. In order to allow them to communicate with the Internet. etc. This translation process is known as Network Address Translation (NAT). Since these addresses cannot be routed over the Internet.0/16.0.168.0.0/8 address space into multiple Class B networks (10.0/16. In the future. should the networks ever be linked (either by a physical connection. you might break the 10.0/24. field office one. computers which use them are not really "part" of the Internet. We will look at NAT in more detail on Page 43.40 To LAN Chapter 3: Network Design To LAN 10.

1.168. host A broadcasts to all hosts the question.1.1. with the corresponding IP addresses 192.3?" When host C sees an ARP request for its own IP address.168.3 Computer A 192. it replies with its MAC address.1. Computer B 192.2 Figure 3. and 192.168.168.2.1.1.168.3 is 00:11:22:aa:bb:cc Computer C 192.1. they must determine each others' MAC addresses.168.2.1.1.1 Hub Computer F: 192.168.2. Computer B: 192..1.168.2 Computer A: 192.168. E.1 00:11:22:aa:bb:cc .1.1.3 Figure 3.3. It is possible to manually configure each host with a mapping table from IP address to MAC address.2 Computer A 192.DATA.6: Computer A needs to send data to 192.Chapter 3: Network Design 41 For two hosts to communicate on a local network.2.2 Computer C 192.168.168. But it must first ask the whole network for the MAC address that responds to 192.168. and F..1.3 Computer D: 192.3. but it is not in the same range as the network above.168.168.3 Computer E: 192.1. but normally the Address Resolution Protocol (ARP) is used to determine this automatically. All three . D.1 who is 192.168. 192.168.168.3? 192.2. Consider now another network with 3 hosts.1.2.168. Computer B 192.2.168.3.1 Hub Computer C: 192. "Who has the MAC address for the IP 192. When using ARP.7: Two separate IP networks. This is another /24 network.168.

1.168.168.168.1 Hub Computer F: 192.2.3 Computer D: 192.168. But what if hosts A.2.1. The first network card uses the IP address 192.4 (host G).4.4. this can be accomplished with the following command: # ip route add 192.4 Computer G 192. Computer B: 192.2 Computer A: 192.2.168.2. In Linux.1. and the other uses 192.0/24 via 192.3 Computer E: 192.1.168. A route tells the OS that the desired network doesn't lie on the immediate link-local network. Host G would then look up host F in its routing table.1.168. and can route packets between them. Notice that the route is added via the IP address on host G that is link-local to the respective network. hosts A-C would add a route via 192.. Now we will add host G. with one plugged into each network.168.2.2. and then sending packets to that MAC address).42 Chapter 3: Network Design hosts can reach each other directly (first using ARP to resolve the IP address into a MAC address.and hosts D-F would add the following: # ip route add 192. and F? They will need to add a route to the other network via host G.4 .168.168. since that IP is not link-local.168.4.2. and C want to reach hosts D. This host has two network cards.168. For example. If host A wants to send a packet to host F.2 Figure 3.1. it would first send it to host G.1. even though it is the same physical machine as 192.4 The result is shown in Figure 3.168.168.4 Computer C: 192. E.1 Hub 192. and it must forward the traffic through the specified router.1.8.4. Host G is now link-local to both networks.8: Host G acts as a router between the two networks.2.168.1. and see that it has a direct . B.168.168.0/24 via 192. Host A could not add a route via 192..

6.9.15.5.Chapter 3: Network Design 43 connection to host F's network.255.6.0 Flags U U UG Metric 0 0 0 Iface eth1 eth0 eth0 Figure 3.5.6. BGP.0 10.0.3 Internal Router eth1 eth0 10. many hops may need to be traversed to reach the ultimate destination.1 Genmask 255. Routes can be updated manually.5.2 10. This is a very simple routing example. 10.15.15.6. but for further reading on the subject.15.255. The default gateway is typically the best route out of your network.0 default Gateway * * 10. OSPF. a host uses the default gateway entry in its routing table. the Internet connection uses one (or more) glob- . usually in the direction of your ISP. see the resources in Appendix A. Since it isn't practical for every machine on the Internet to know the route to every other.0 255. Network Address Translation (NAT) In order to reach hosts on the Internet. or can dynamically react to network outages and other events. Configuring dynamic routing is beyond the scope of this book.15. and OLSR. the packet is forwarded to its default gateway.0.4 10. or NAT. host G would resolve the hardware (MAC) address of host F and forward the packet to it.0 0. On a NAT router. When a router receives a packet destined for a network for which it has no explicit route. A NAT device is a router that manipulates the addresses of packets instead of simply forwarding them.5. publicly routable IP addresses.9: When no explicit route exists to a particular destination. Some examples of popular dynamic routing protocols are RIP.1 Routing table for internal router: Destination 10. As networks get more complex.15. where the destination is only a single hop away from the source.3 10.15.15. Finally. RFC1918 addresses must be converted to global. This is achieved using a technique known as Network Address Translation.15.255.255.6. we make use of a routing entry known as the default route (also known as the default gateway).2 Internet 10. An example of a router that uses a default gateway is shown in Figure 3.

As far as the network users can tell.1. To 10. While NAT performs some firewall-like functions.1.2 10. The major consequence of using NAT is that machines from the Internet cannot easily reach servers within the organization without setting up explicit forwarding rules on the router.2. although some applications (such as Voice over IP and some VPN software) can have difficulty dealing with NAT.1. they are directly connected to the Internet and require no special software or drivers.1.44 Chapter 3: Network Design ally routed IP addresses.235. The NAT router allows the global address(es) to be shared with all of the inside users. it is not a replacement for a real firewall. The NAT router translates outbound packets to use the global IP address as they leave the network. RFC1918 addresses should be filtered on the edge of your network to prevent accidental or malicious RFC1918 traffic entering or leaving your network.226 ? Internet 192. Connections initiated from within the private address space generally have no trouble.1. and translates them back again as they are received from the Internet. this can be considered a bug (since it makes it harder to set up two-way communication) or a feature (since it effectively provides a "free" firewall for your entire organization). while the private network uses an IP address from the RFC1918 private address range.4 Figure 3.1. .1.0.1 10.3 10.90.3 69.1 NAT router 10.10: Network Address Translation allows you to share a single IP address with many internal hosts.1. who all use private addresses. It converts the packets from one form of addressing to the other as the packets pass through it. Depending on your point of view. They simply use the NAT router as their default gateway. but can make it difficult for some services to work properly.1. and address packets as they normally would.1.

. these protocols (and others) are known as the Internet Protocol Suite. number 6). By now you should have a solid understanding of how computers on the network are addressed. UDP is designed for connectionless streams of information. and still be distinguished from each other. and 80 as the destination port. notify the sender of another packet of a possible routing loop (time exceeded). 25 for email). and does not guarantee delivery at all. or in any particular order. it has message types. There are a number of protocols that are run in conjunction with IP that provide features as critical to normal operations as IP itself. used to reach well known services such as email and web servers. When sending a response to such packets. Different message types are used to request a simple response from another computer (echo request). For example. but it must connect to the proper port on the server (e. 80 for web. web servers normally listen on TCP port 80. The most commonly used protocols are the Transmission Control Protocol (TCP. Every packet has a source and destination port number. even when separated by many intermediary machines. retries. which are also numbers. we mean that it will accept packets that use its IP as the destination IP address. Now let's take a brief look at the physical hardware that implements these network protocols. Every packet specifies a protocol number which identifies the packet as one of these protocols.). Some port numbers are well defined standards. packet reordering and reassembly. Servers usually do not care about the source IP or source port. Taken as a group. or inform the sender that a packet that could not be delivered due to firewall rules or other problems (destination unreachable). Rather than port numbers. the server will use its own IP as the source IP. or simply TCP/IP for short. etc. although sometimes they will use them to establish the identity of the other side. TCP is a session oriented protocol with guaranteed delivery and transmission control features (such as detection and mitigation of network congestion. number 1). and 80 as the source port. Port numbers allow multiple services to be run on the same IP address.Chapter 3: Network Design 45 Internet Protocol Suite Machines on the Internet use the Internet Protocol (IP) to reach each other. it may use any source port number on its side which is not already in use. number 17). The ICMP protocol is designed for debugging and maintenance on the Internet. When we say that a service "listens" on a port (such as port 80). When a client connects to a service. and the Internet Control Message Protocol (ICMP. The TCP and UDP protocols introduce the concept of port numbers. and how information flows on the network between them. and SMTP email servers listen on TCP port 25. User Datagram Protocol (UDP.g.

where misbehaving devices can disable the entire segment. the scope of a MAC address is limited to a broadcast domain. The official standard is called IEEE 802. and both must back off and retransmit their packets later. switches. ADSL modem. which is defined as all the computers connected together by wires. Its function is like that of an IP address. or wireless device. and each host remains responsible for detecting collisions during transmission. Hub-based networks are generally more robust than coaxial Ethernet (also known as 10base2 or ThinNet). The network topology is a star. It is sometimes used to connect individual computers to the Internet. But hubs are limited in their usefulness. they corrupt each other's transmissions. some hubs can disconnect (partition) that port for a while to limit its impact on the rest of the network. the medium which was once supposed to carry light waves through free space. assigned by the manufacturer of the network card. This defines a data rate of 100 megabits per second. and bridges. and end nodes (devices and additional switches) at the edges. since they can easily become points of congestion on busy networks. They repeat the signals received by each port out to all of the other ports. via a router. only one port can successfully transmit at a time. if you connect a single computer to the Internet. and are not transmitted across routers.3. and retransmitting its own packets when needed. While a port is partitioned. The name comes from the physical concept of the ether. since it serves as a unique identifier that enables devices to talk to each other. Due to this design. When problems such as excessive collisions are detected on a port. MAC addresses are never used directly on the Internet. devices attached to it cannot communicate with the rest of the network. you may not use Ethernet at all. with modular RJ-45 connectors on the end. Hubs can therefore be considered to be simple repeaters. but not crossing routers or Internet gateways. However. They work at the physical layer (the lowest or first layer). This is known as a collision. MAC addresses Every device connected to an Ethernet network has a unique MAC address.46 Chapter 3: Network Design Ethernet Ethernet is the name of the most popular standard for connecting together computers on a Local Area Network (LAN). with switches or hubs at the center of each star. running over twisted pair wires. However. Hubs Ethernet hubs connect multiple twisted-pair Ethernet devices together. . The most common Ethernet standard is called 100baseT. If two devices transmit at the same time. hubs.

it makes a note of the source MAC address. It stores this information in an internal MAC table. If the destination MAC address is not found in the MAC table. the packet is filtered and is not forwarded. the switch determines which ports are communicating directly and temporarily connects them together. Overall performance is slower. for example.Chapter 3: Network Design 47 Switches A switch is a device which operates much like a hub. When a packet arrives at a port on a switch. which it associates with that port. Rather than repeating all traffic on every port. allow the creation of VLANs. More expensive switches can switch traffic by inspecting packets at higher levels (at the transport or application layer). old network hubs should be replaced whenever possible with new switches. the packet is then sent to all of the connected interfaces. The switch then looks up the destination MAC address in its MAC table. Switches work at the data link layer (the second layer). when you want to explicitly allow a monitoring machine to see all of the traffic on the network. Switches create virtual connections between receiving and transmitting ports. and are replacing them in many situations. Hubs vs. This simplicity introduces both a performance penalty and a security issue. and implement other advanced features. Switches generally provide much better performance than hubs. . since they interpret and act upon the MAC address in the packets they receive. Since all traffic is seen by all ports. since the available bandwidth must be shared between all ports. and transmits the packet on the matching port. If the destination port matches the incoming port. but provides a dedicated (or switched) connection between ports. since they inefficiently rebroadcast all traffic on every port. Switches Hubs are considered to be fairly unsophisticated devices. especially on busy networks with many computers. They are not much more expensive than hubs. However. Most switches provide monitor port functionality that enables repeating on an assigned port specifically for this purpose. the price of switches have reduced dramatically over the years. This yields better performance because many virtual connections can be made simultaneously. A hub can be used when repetition of traffic on all ports is desirable. Hubs were once cheaper than switches. Therefore. any host on the network can easily monitor all of the network traffic.

Routers can be dedicated hardware devices (such as Cisco or Juniper routers) or they can be made from a standard PC with multiple network cards and appropriate software.1. DSL. they have one connection to each network.4 Switch 10. A managed switch that provides upload and download byte counts for every physical port can greatly simplify network monitoring.1.2 10.3 10. full or half duplex) per port. Routers sit at the edge of two or more networks. 1000baseT.1. By definition. or they may be accessed via telnet.4 to: 10.1.1.2 10. and usually include port counters for easy bandwidth accounting. and as border machines they may take on other responsibilities as well as routing. Many routers have firewall capabilities that provide a mechanism to filter or redirect packets that do not fit security or . such as Ethernet. 100baseT.48 Chapter 3: Network Design to: 10. These services are typically available via SNMP.1. It may include support for different types of network media.4 Figure 3. Routers and firewalls While hubs and switches provide connectivity on a local network segment. or dial-up.11: A hub simply repeats all traffic on every port.1. ssh. ATM. or a custom configuration tool. enable triggers to watch for network events (such as changes in MAC address or malformed packets).1.1.1.1.3 10. dedicated connection between the ports that need to communicate. a web interface. Some of these services include the ability to set the link speed (10baseT.1. a router's job is to forward packets between different network segments. Both hubs and switches may offer managed services.1.1. while a switch makes a temporary.1. A router typically has two or more physical network interfaces.1.4 Hub 10.

and VSAT terminals terminate at an Ethernet jack.12: Many DSL modems. often with NAT functionality. although the least expensive devices often omit this feature. and more involved configuration. Each physical network has an associated piece of terminal equipment. and higher reliability than software routers on PCs. wireless access points. They tend to have much better performance. Routers vary widely in cost and capabilities.Chapter 3: Network Design 49 access policy requirements. The lowest cost and least flexible are simple. which consists of an operating system running on a standard PC with multiple network interfaces. However. For example. The most expensive devices are high-end dedicated hardware routers. CSU/DSUs. cable modems. Standard operating systems such as Microsoft Windows. Linux. dedicated hardware devices. Most modern routers offer mechanisms to monitor and record performance remotely. used to share an Internet connection between a few computers. It is also possible to purchase technical support and maintenance contracts for them. a large number of complex and potentially unreliable parts. with high power consumption. Other equipment DSL Modem Ethernet Switch Cable Modem Ethernet Switch CSU/DSU Ethernet Switch VSAT Terminal Ethernet Switch Figure 3. they suffer from the same problems as conventional PCs. usually via the Simple Network Management Protocol (SNMP). made by companies like Cisco and Juniper. and BSD are all capable of routing. more features. and are much more flexible than the low-cost hardware devices. The next step up is a software router. They may also provide Network Address Translation (NAT) services. VSAT connections consist of a satellite dish connected to a termi- .

3 Bobʼs computer: 192. Since your Internet connection ultimately comes from your ISP.54 192.91.6.50 Chapter 3: Network Design nal that either plugs into a card inside a PC. they can send data packets to the IP address of any other node. Bob! Hello. DSL lines use a DSL modem that bridges the telephone line to a local device.231. while others may not.38. Some provide mechanisms for monitoring performance. Each network segment has a router with two IP addresses. but nearly always end at an Ethernet jack.16.6.231.8.17.17. or to an internal PC card bus. In this example.15. Through the use of routing and forwarding.1 Router 2.33 10.1 23.15.205 216.1 Router 216. Putting it all together Internet Router 172.17. making it “link local” to two different networks.5 Figure 3. usually via a plug-in card or serial port.154 172. these packets can reach nodes on networks that are not physically connected to the originating node. Packets are forwarded between routers until they reach their ultimate destination. This process describes much of what “happens” on the Internet. you can see the path that the packets take as Alice chats with Bob using an instant messaging service. or ends at a standard Ethernet connection. Some kinds of telecom circuit (such as a T1 or T3) use a CSU/DSU to bridge the circuit to a serial port or Ethernet. either an Ethernet network or a single computer via USB.13: Internet networking.168. And there are many different kinds of wireless networking equipment that connect to a variety of radios and antennas. Once all network nodes have an IP address. Cable modems bridge the television cable to Ethernet.8. Alice! Aliceʼs computer: 10.16. Standard dialup lines use modems to connect a computer to the telephone.38. you should follow their recommendations when choosing equipment that bridges their network to your Ethernet network. Each dotted line represents an .1 Hello.41.41.168. The functionality of these devices can vary significantly between manufacturers.

After all. or any other kind of physical network. One side of a point-to-point link will have an Internet connection. Neither Alice nor Bob need to be concerned with how those networks operate.14: A point-to-point link allows a remote site to share a central Internet connection. If it weren t for Internet protocols and the cooperation of everyone on the net. reliable point-to-point links in excess of thirty kilometers are possible. Designing the physical network It may seem odd to talk about the “physical” network when building wireless networks. but cannot afford such a connection for an important building just off campus. . If the main building has an unobstructed view of the remote site. a point-to-point connection can be used to link the two together. wireless networks are naturally arranged in these three logical configurations: point-to-point links. The cloud symbol is commonly used to stand in for “The Internet”. where is the physical part of the network? In wireless networks. a university may have a fast frame relay or VSAT connection in the middle of campus.Chapter 3: Network Design 51 Ethernet cable. This can augment or even replace existing dial-up links. and represents any number of intervening IP networks. this kind of communication would be impossible. How do you arrange the equipment so that you can reach your wireless clients? Whether they fill an office building or stretch across many miles. While different parts of your network can take advantage of all three of these configurations. a wireless link. and multipoint-to-multipoint clouds. Point-to-point Point-to-point links typically provide an Internet connection where such access isn t otherwise available. the physical medium we use for communication is obviously electromagnetic energy. For example. pointto-multipoint links. as long as the routers forward IP traffic towards the ultimate destination. any individual link will fall into one of these topologies. With proper antennas and clear line of sight. But in the context of this chapter. while the other uses the link to reach the Internet. Poin t to poin t lin k VSAT Figure 3. the physical network refers to the mundane topic of where to put things.

computers. You could connect the site with a point-to-point link. even if there is no direct connection to the Internet. Omnidirectional antenna VSAT Figure 3. All three sites can also communicate directly at speeds much faster than VSAT. Suppose you have to physically drive to a remote weather monitoring station. The laptops do not communicate with each other directly. Rather than setting up several point-to-point links to distribute the Internet connection. The typical example of a point-tomultipoint layout is the use of a wireless access point that provides a connection to several laptops. this is a point-to-multipoint application.52 Chapter 3: Network Design Of course. Suppose the remote building on top of the hill is connected to the central campus with a point-to-point link. Access points. Point-to-multipoint networking can also apply to our earlier example at the university. and laptops are all examples of nodes.15: The central VSAT is now shared by multiple remote sites. another node can join the network and make use of the central Internet connection. . This is a classic 2. but must be in range of the access point in order to use the network. high in the hills. in order to collect the data which it records over time. routers. Whenever several nodes2 are talking to a central point of access. it may be able to see other important locations that can t be seen directly from the central campus. once a single point-to-point connection has been made. A node is any device capable of sending and receiving data on a network. Point-to-point links don t necessarily have to involve Internet access. Point-to-multipoint The next most commonly encountered network layout is point-tomultipoint. Wireless networks can provide enough bandwidth to carry large amounts of data (including audio and video) between any two points that have a connection to each other. without the need to actually travel to the site. allowing data collection and monitoring to happen in realtime. By installing another point-to-point link at the remote site. more can be used to extend the network even further. If the remote building in our example is at the top of a tall hill. a single antenna could be used that is visible from several remote buildings.

since every participant potentially carries the traffic of every other. then that connection can be shared among all of the clients. which will be addressed later in this chapter. which means that they automatically detect routing problems and fix them as needed. Extending a mesh network is as simple as adding more nodes. The benefit of this network layout is that even if none of the nodes are in range of a central access point. Multipoint-to-multipoint The third type of network layout is multipoint-to-multipoint. Every node on the network carries the traffic of every other as needed. two-way data networks behave very differently than broadcast radio. they can still communicate with each other. Multipoint-to-multipoint networks tend to be difficult to troubleshoot. Multipoint-tomultipoint clouds typically have reduce capacity compared to point-to-point or point-to-multipoint networks. VSAT Figure 3. but don t make the classic mistake of installing a single high powered radio tower in the middle of town and expecting to be able to serve thousands of clients. Two big disadvantages to this topology are increased complexity and lower performance. as you would with an FM radio station. Note that there are a number of performance issues with using point-tomultipoint over very long distance. and all nodes communicate with each other directly. . which is also referred to as an ad-hoc or mesh network. Good mesh network implementations are self-healing. As we will see. due to the additional overhead of managing the network routing and increased contention in the radio spectrum. there is no central authority. If one of the nodes in the "cloud" happens to be an Internet gateway. Every point can reach each other at very high speed. Such links are possible and useful in many circumstances.16: A multipoint-to-multipoint mesh. or use the central VSAT connection to reach the Internet. In a multipoint-to-multipoint network. Security in such a network is also a concern. due to the large number of changing variables as nodes join and leave the network.Chapter 3: Network Design 53 example of a wide area point (remote site on the hill) to multipoint (many buildings in the valley below) connection.

then they are ready to negotiate data link layer connectivity. allowing the network to spread organically between laptop users who all ultimately use the original point-to-point link to access the Internet.11b radio card is set to channel 2 while another is set to channel 11. then the radios cannot communicate with each other.54 Chapter 3: Network Design Nevertheless. As we saw in Chapter 2. mesh networks are useful in many circumstances. Master mode (also called AP or infrastructure mode) is used to create a service that looks like a traditional access point. The wireless card creates a network with a specified name (called the SSID) and channel. Use the technology that fits All of these network designs can be used to complement each other in a large network. While in master mode. 802. wireless cards manage all communications related to the network (authenticating wire- .11a device cannot interoperate with an 802. to use a long distance wireless link to provide Internet access to a remote location.11 wireless networks Before packets can be forwarded and routed to the Internet. wireless cards must agree on a common channel. Now that we have a clear idea of how wireless networks are typically arranged.11a radios at around 5 GHz.11a/b/g device can operate in one of four possible modes: 1.11b/g radios at around 2.11a radios will talk to 802. But an 802. It is a common practice. network nodes cannot talk to each other and route packets. To provide physical connectivity.4 GHz. layers one (the physical) and two (the data link) need to be connected. When two wireless cards are configured to use the same protocol on the same radio channel. and 802. we can begin to understand how communication is possible over such networks.11b/g device. this means that 802. Each 802. and then set up an access point on the remote side to provide local wireless access. wireless network devices must operate in the same part of the radio spectrum. More specifically. and can obviously make use of traditional wired networking techniques whenever possible. since they use completely different parts of the electromagnetic spectrum. We will see an example of how to build a multipoint-to-multipoint mesh network using a routing protocol called OLSR later in this chapter. One of the clients of this access point may also act as a mesh node. Without link local connectivity.11b/g radios will talk to other 802. for example. If one 802. and offers network services on it.

and if those credentials are accepted. Clients. They then present any necessary credentials to the master. and will automatically change their channel to match it. Managed mode is sometimes also referred to as client mode. wireless cards transmit no data. Ad-Hoc node cannot talk to a Client node Ad-Hoc X Client (Managed) Client (Managed) AP cannot talk to another AP Access Point (Master) Ad-Hoc Access Point (Master) X X Ad-Hoc node cannot talk to an Access Point Client (Managed) Client node cannot talk to another Client node Client (Managed) X Client (Managed) Ad-Hoc Ad-Hoc Ad-Hoc Ad-Hoc Ad-Hoc Ad-Hoc Ad-Hoc Ad-Hoc nodes can only talk to other Ad-Hoc nodes that are in range Figure 3. 2. In a multipoint-to-multipoint mesh. and will only communicate with an associated master. Wireless cards in managed mode will join a network created by a master. When in monitor mode. while the other(s) operate in managed mode. Managed mode cards do not communicate with each other directly. Monitor mode is used by some tools (such as Kismet. handling channel contention. the radios all operate in ad-hoc mode so that they can communicate with each other directly. . they are said to be associated with the master. Monitor mode is not used for normal communications. 3. When implementing a point-to-point or point-to-multipoint link. each wireless card communicates directly with its neighbors. and Ad-Hoc nodes. etc. Nodes must be in range of each other to communicate.) Wireless cards in master mode can only communicate with cards that are associated with it in managed mode. In ad-hoc mode.Chapter 3: Network Design 55 less clients. see Chapter 6) to passively listen to all radio traffic on a given channel. and must agree on a network name and channel. 4. Ad-hoc mode creates a multipoint-to-multipoint network where there is no single master node or AP. repeating packets. one radio will typically operate in master mode.17: APs. This is useful for analyzing problems on a wireless link or observing spectrum usage in the local area.

in our example both computers could achieve full speed when operating ad-hoc. In ad-hoc mode there is no hierarchical master-client relationship.for example. attached to a DSL line or other large scale wired network. ad-hoc is more flexible but has a number of performance issues as compared to using the master / managed modes. which operate in managed mode. The more nodes. if two clients A and C can't directly “see” each other with their wireless interfaces. the effective speed between both clients will be only 300 kByte/sec. A single transmission may have a speed of 600 kByte/sec (thats about the maximum speed you could achieve with 802. Thus. WiFi cards in managed mode can't communicate directly.thus.the base station then sends data back to the phone of your friend. As we will see later in this chapter. Ad-hoc nodes do not repeat by default. Mesh networking with OLSR Most WiFi networks operate in infrastructure mode . Any traffic between clients connected to an access point has to be sent twice. so it is likely that you will want to run a high repeater site in master or ad-hoc mode.have to use the access point as a relay. This topology is similar to a mobile phone (GSM) service. two laptops on the same table . they still can communicate as long as the AP is in the wireless range of both clients. Mesh networks are based on the strategy that every mesh-enabled node acts as a relay to extend coverage of the wireless network. because the data has to be repeated by the access point before it reaches its target. Clients . The disadvantage to ad-hoc mode is that clients do not repeat traffic destined for other clients. Remember that managed mode clients cannot communicate with each other directly. under ideal circumstances. In such a hotspot the access point usually acts as a master station that is distributing Internet access to its clients. but they can effectively do the same if routing is applied.they consist of an access point somewhere (with a radio operating in master mode). Mobile phones connect to a base station . . client A sends data to the access point B. either. If you make a joke call to a friend that is sitting on the other side of the table. If client A and C communicate. and then the access point will retransmit the data to client C.without the presence of such a base station mobiles can't communicate with each other. the better the radio coverage and range of the mesh cloud. Nodes can communicate directly as long as they are within the range of their wireless interfaces.11b) in our example .56 Chapter 3: Network Design It is important to keep these modes in mind when designing your network layout. your phone sends data to the base station of your provider that may be a mile away . In the access point example.

Unfortunately. If an ad-hoc network consists of only a few nodes that are up and running at all time. Nodes can fail. And no one wants to update several routing tables by hand if one node is added to the network. don't move and always have stable radio links . 2000. Also. XP. There is one big tradeoff that must be mentioned at this point. and interference can make radio links unusable at any time. Windows 98.it is possible to write individual routing tables for all nodes by hand. Client (A) Access Point (B) Client (C) 57 In the same setting. Ad-Hoc (A) Ad-Hoc (B) Ad-Hoc (C) X Figure 3. Ad-Hoc nodes A and C can communicate with node B. cheap ad-hoc mesh networks can provide good radio coverage on the last mile(s) of a community wireless network at the cost of speed-.especially if the density of nodes and transmit power is high. Mesh routing with olsrd The Optimized Link State Routing Daemon . By using routing protocols that automatically maintain individual routing tables in all nodes involved.org is a routing application developed for routing in wireless networks. Access Point B will relay traffic between the two nodes. we can avoid these issues. Linux. but not with each other. It is a open-source project that supports Mac OS X. WiFi enabled devices roam around.from olsr. Thus. FreeBSD. We will concentrate on this routing software for several reasons. In Ad-Hoc mode.olsrd . those conditions are rarely met in the real world. the available bandwidth is significantly reduced every time traffic is repeated by intermediate nodes on the way from A to B. there will be interference in transmission due to nodes sharing the same channel.Chapter 3: Network Design Clients A and C are in range of Access Point B but not each other.18: Access point B will relay traffic between clients A and C. OpenBSD and . Popular routing protocols from the wired world (such as OSPF) do not work well in such an environment because they are not designed to deal with lossy links or rapidly changing topology. If the device only uses one radio interface. node B will not relay traffic between A and C by default.a long list of ifs .

but there is no reason to disable Link Quality Extension unless compliance with other implementations is required. and ships standard on Metrix kits running Pyramid.58 Chapter 3: Network Design NetBSD. since the routing table has to be updated less often. There are pros and cons to proactive routing.depending on how many interfaces are involved and how often the routing tables are updated.org can be configured to behave according to the IETF draft that lacks this feature . Based on practical experience of the free networking community.org started as a master thesis of Andreas Toennesen at UniK University. Higher protocol traffic overhead and more CPU load are among the disadvantages. Olsrd is available for access points that run Linux like the Linksys WRT54G. reactive routing algorithms seek routes only when it is necessary to send data to a specific node. Olsrd can handle multiple interfaces and is extensible with plug-ins. Olsrd now differs significantly from the original draft because it includes a mechanism called Link Quality Extension that measures the packet loss between nodes and calculates routes according to this information. which began as an IETF-draft written at INRIA France. In Berlin. Maintaining routes in a mesh cloud with static nodes takes less effort than a mesh with nodes that are constantly in motion. Mechanism A node running olsrd is constantly broadcasting 'Hello' messages at a given interval so neighbors can detect it's presence. Note that there are several implementations of Optimized Link State Routing. This extension breaks compatibility to routing daemons that follow the INRIA draft. The olsrd available from olsr. Each node maintains a routing table covering the whole mesh cloud. There is clearly a limit to what extent a proactive protocol can scale . The biggest advantage of proactive routing is that you know who is out there and you don't have to wait until a route is found. The average CPU load caused by olsrd on a Linksys WRT54G running at 200 MHz is about 30% in the Berlin mesh. It supports IPv6 and it is actively developed and used by community networks all over the world. a node knows about the existence of every other node in the mesh cloud and which nodes may be used to route traffic to them. In contrast. Asus Wl500g. Every node computes a statistic how many 'Hellos' have been lost or received from each neighbor - . This approach to mesh routing is called proactive routing. AccessCube or Pocket PCs running Familiar Linux. the Freifunk community is operating a mesh cloud where olsrd has to manage more than 100 interfaces. Theory After olsrd is running for a while. and there are many other ideas about how to do mesh routing that may be worth mentioning. the routing daemon was modified. The implementation from olsr.

A DHCP request will not be answered by a DHCP server if the node requesting DHCP needs a multihop link to connect to it. Installation packages are available for OpenWRT. Practice Olsrd implements IP-based routing in a userland application . However. please read the documentation that is shipped with the source package. Note that multipoint relays are only chosen for the purpose of forwarding TC messages. First of all. Mac OS X. If everything is configured properly all you have to do is start the olsr program. a dynamic gateway plugin was written. If not. When a mesh node roams around it will detect gateways into other networks and always choose the gateway that it has the best route to. Debian GNU/Linux and Windows. Such transmissions are redundant if a node has many neighbors. HNA messages make olsrd very convenient when connecting to the Internet with a mobile device. unnecessary overhead can be generated. This problem could be solved by using IPv6. To overcome this problem. AccessCube. olsrd is by no means bullet proof. If you have to compile from source. Thus. The concept of multipoint relays is a new idea in proactive routing that came up with the OLSR draft. If every node rebroadcasts topology information that it has received. Two other message types exist in OLSR that announce information: whether a node offers a gateway to other networks (HNA messages) or has multiple interfaces (MID messages). OLSR is a standard part of Metrix Pyramid. olsrd ceases to send false HNA messages. The pseudo-gateway is a black hole. an olsrd node decides which neighbors are favorable multipoint relays that should forward its topology control messages. since there is plenty of space available to generate a unique IP from the MAC address of each card involved (as suggested in "IPv6 State- . There is not much to say about what this messages do apart from the fact that they exist. If a node announces that it is an Internet gateway which it isn't because it never was or it is just offline at the moment . and applying dhcp relay throughout a mesh is likely impractical.installation is pretty easy. It is not recommended (nor practicable) to use DHCP in an IP-based mesh network. it must be ensured that every node has a unique statically assigned IP-Address for each interface used for the mesh.the other nodes will nevertheless trust this information. The plugin will automatically detect at the gateway if it is actually connected and whether the link is still up. It is highly recommended to build and use this plugin instead of statically enabling HNA messages. Payload is routed considering all available nodes. The gained topology information is broadcasted as topology control messages (TC messages) and forwarded by neighbors that olsrd has chosen to be multipoint relays.Chapter 3: Network Design 59 thereby gaining information about the topology and link quality of nodes in the neighborhood.

They may be unable to connect to other devices on the same table. A wiki-page where every interested person can choose an individual IPv4 address for each interface the olsr daemon is running on may serve the purpose quite well.457 GHz Cell: 02:00:81:1E:48:10 Bit Rate:2 Mb/s Sensitivity=1/3 Retry min limit:8 RTS thr=250 B Fragment thr=256 B Encryption key:off Power Management:off Link Quality=1/70 Signal level=-92 dBm Noise level=-100 dBm Rx invalid nwid:0 Rx invalid crypt:28 Rx invalid frag:0 Tx excessive retries:98024 Invalid misc:117503 Missed beacon:0 It is important to set the 'Request To Send' threshold value RTS for a mesh. You can check this out with the command iwconfig when using GNULinux.11 standard for ad-hoc networking and may fail miserably to connect to a cell. It just has to be ensured that settings are the same everywhere. WiFi cards made by Intel that are shipped with Centrino Notebooks are notorious for doing this. Make sure the interface joins the same wireless channel.255.org mode ad-hoc channel 10 rts 250 frag 256 Verify that the wireless part of the WiFi card has been configured so it has an ad-hoc connection to other mesh nodes within direct (single hop) range. When a default olsrd configuration file is issued.255. since olsrd can be configured to override the broadcast addresses with this default. They may even confuse other cards that behave according to the standard by creating their own Cell-ID on the same channel with the same wireless network name. uses the same wireless network name ESSID (Extended Service Set IDentifier) and has the same Cell-ID as all other WiFi-Cards that build the mesh. Here is an example command how to configure a WiFi card with the name wlan0 using Linux: iwconfig wlan0 essid olsr. this feature should be enabled to avoid confusion of the kind "why can't the other nodes see my machine?!?" Now configure the wireless interface.org" Mode:Ad-Hoc Frequency:2.11b ESSID:"olsr. Zitterbart. Weniger and M. There is just not an easy way to automate the process if IPv4 is used. 2002). There will be collisions on the radio channel between the transmissions of . There is no reason to enter the broadcast address explicitly. Here is the output on my machine: wlan0 IEEE 802. even if they are set up with the correct channel and wireless network name. Olsrd can do this on its own.255 on mesh interfaces in general as a convention.60 Chapter 3: Network Design less Address Autoconfiguration in large mobile ad hoc networks" by K. Many WiFi cards or their respective drivers do not comply with the 802. The broadcast address should be 255.

A simple olsrd.here set to 256 . so it must be smaller than the IP packet size.3" { PlParam "Interval" "60" PlParam "Ping" "151.Chapter 3: Network Design 61 nodes on the same wireless channel.conf It is not practical to provide a complete configuration file here. but in a noisy environment this reduces the error penalty and allows packets to get through interference bursts. Fragmentation allows to split an IP packet in a burst of smaller fragments transmitted on the medium.255.1. On BSD and Linux the configuration file /etc/olsrd. TCP is very sensitive to collisions.so. and RTS will mitigate this.like announcing black holes.255 } . the configuration file of olsrd must be altered in order that olsrd finds and uses the interfaces it is meant to work on.255.0. The RTS threshold value must be smaller than the IP-Packet size and the 'Fragmentation threshold' value .1" PlParam "Ping" "194. Unfortunately this tempts users that lack background knowledge to do stupid things .25.conf has to be edited with a text editor. For Mac OS-X and Windows there are nice GUI's for configuration and monitoring of the daemon available. This adds overhead.a value equal to the maximum IP packet size disables the mechanism. RTS/CTS adds a handshake before each packet transmission to make sure that the channel is clear. but increases performance in case of hidden nodes . This parameter sets the maximum size before a data packet is split and sent in a burst . This adds overhead. Mesh networks are very noisy because nodes use the same channel and therefore transmissions are likely to interfere with each other.2. These are some essential settings that should be checked. so it is important to switch RTS on.1.otherwise it will be disabled. UseHysteresis TcRedundancy MprCoverage LinkQualityLevel LinkQualityWinSize no 2 3 2 20 LoadPlugin "olsrd_dyn_gw.129" } Interface "ath0" "wlan0" { Ip4Broadcast 255. Once a valid IP-address and netmask is assigned and the wireless interface is up.and hidden nodes are the default in a mesh! This parameter sets the size of the smallest packet (in bytes) for which the node sends RTS. Setting fragmentation threshold is recommended.

000 ILQ 1.00 --.1 192.168.000 1.1:1.000 1.168.000 lost 0 0 total 20 20 NLQ 1.120.168. because debugging creates a lot of CPU load.conf.00 Using OLSR on Ethernet and multiple interfaces It is not necessary to have a wireless interface to test or use olsrd .LINKS IP address 192.51 --------------------------------------------.1 192. .120.000 0.19:27:45.168.17 LQ 1.000 LQ 1.120.168.120.00 (one-hop) 192.DIJKSTRA 192.168.although that is what olsrd is designed for.168. especially for the first time.000 1.00 (one-hop) --. the driver situation is improving now.000 SYM YES YES MPR NO NO MPRS YES YES will 3 6 --.51 -----------------------------------------------.120. Many WiFi cards and drivers are buggy in ad-hoc mode.00 1.3 Dest IP addr 192.000 1.168.000 1. After these steps have been done.168.120.120. so the implementation of the ad-hoc mode was done sloppily by many manufacturers.000 NLQ 1.19:27:45.19:27:45.000 ETX 1.3 LQ 1. You can see what olsrd does and monitor how well the links to your neighbors are.17 192.120. WiFiinterfaces don't have to operate always in ad-hoc mode to form a mesh when mesh nodes have more than one interface. but these basic options should get you started. Ad-hoc mode has not had many users so far.168.00 1. It may as well be used on any NIC.000 ETX 1.TOPOLOGY Source IP addr 192.3:1. For dedicated links it may be a very good option to have them running in infrastructure mode.3 hyst 0.62 Chapter 3: Network Design There are many more options available in the olsrd.19:27:45.because everybody expects at least this feature to work. olsrd can be started with a simple command in a terminal: olsrd -d 2 I recommend to run it with the debugging option -d 2 when used on a workstation. With the rising popularity of mesh networks.51 --------------------------------------------.51 -------------------------------------------.NEIGHBORS IP address 192.1 192. but infrastructure mode works fine .120.120. The output should look something like this: --. On embedded devices the debug level should be 0 (off).000 1.

254. Here a little HOWTO for the network topology visualization plugin olsrd_dot_draw.26. olsrd_dot_draw outputs the topology in the dot file format on TCP port 2004.11 169. In fact the idea of having a protocol that does everything for newbies that want to have a small to medium sized routed LAN is very appealing.but it works.89 1.00 169.00 1.org website for a complete list.254.0 10.00 1.06 169.Chapter 3: Network Design 63 Many people use olsrd on wired and wireless interfaces .25. They just connect antennas to their WiFi cards.18 1.2.37. This should not be taken as a recommendation .11 1.1 Figure 3.13 1.why not? They expect olsrd to solve every networking problem.11 1. Installing the dot_draw Plugin Compile the olsr plugins separately and install them.118 HNA 1.00 1.19: An automatically generated OLSR network topology.45 169.2 HNA 1.15.161 1.00 10.39 1.00 1.25.254.00 1.3 1.06 1.00 HNA 1.but .4.it is just amazing what people do with such a protocol and that they have such success with it.15.15.25 1.42 1. Check out the olsr.1 10.00 10.11 1.3 1.00 0.00 1.1 1.0.00 1.254.23.11 HNA 10.00 1.00 10.2.4 1. enable olsrd to run on all computers and all interfaces and fire it up.11 1.06 1.254.0/0.00 1.0. The graphviz tools can then be used to draw the graphs.3. . The parameter "port" specifies the TCP port.0. That is quite an abuse of a protocol that was designed to do wireless networking on lossy links .13. Clearly it is not necessary to send 'Hello' messages on a wired interface every two seconds .0. Often it is very good for the understanding of a mesh network to have the ability to show the network topology graphically.3.00 1.11 1.15.243.23.conf.13 1. connect cables to their Ethernet cards.254.13 3.they don't think about network architecture. Plugins A number of plugins are available for olsrd. 169. The parameter "accept" specifies which host is accepted to view the Topology Information (currently only one) and is "localhost" by default.13.2 169. To load the plugin add the following lines to /etc/olsrd.36 1.

org/ Download the script at: http://meshcube.pl and view the topology updates in near-realtime.5" PlParam "port" "2004" } Then restart olsr and check if you get output on TCP Port 2004 telnet localhost 2004 After a while you should get some text output.org/nylon/utils/olsr-topology-view. Troubleshooting As long as the WiFi-cards can 'see' each other directly with their radios. doing a ping will work whether olsrd is running or not. then the IP-addresses.pl Now you can start the script with ./olsr-topology-view. but olsrd doesn't find routes.0.168. This works because the large netmasks effectively make every node link-local. so routing issues are sidestepped at the first hop.graphviz. Now you can save the output graph descriptions and run the tools dot or neato form the graphviz package to get images. Finally. .so. are you running a firewall? Make sure it doesn't block UDP port 698. Most headaches people face with WiFi in Ad-Hoc mode are caused by the fact that the ad-hoc mode in drivers and cards are implemented sloppily. If it is not possible to ping nodes directly when they are in range it is most likely a card/driver issue.3" { PlParam "accept" "192.org/ • ImageMagick. or your network settings are wrong. If the machines can ping each other. First install the following packages on your workstation: • graphviz. This should be checked first if things do not seem to work as expected.0. Bruno Randolf has written a small perl script which continuously gets the topology information from olsrd and displays it using the graphviz and ImageMagick tools. netmask and broadcast address should be checked. http://www. http://www.64 Chapter 3: Network Design LoadPlugin "olsrd_dot_draw.imagemagick.

If users attempt to push more than 22 megabits through the link. long downloads. Large attachments. such as VSAT. a single 802. so a fair amount of lag can be tolerated.Chapter 3: Network Design 65 Estimating capacity Wireless links can provide significantly greater throughput to users than traditional Internet connections. 22 Mbps means that in any given second. etc. As mentioned earlier. not as email. it will take longer than one second. or DSL. email is asynchronous and intermittent. up to 22 megabits can be sent from one end of the link to the other. and how they use the wireless link. Various Internet applications require different amounts of throughput. it is put in a queue. This backlog of data increases the time needed for the most recently queued bits to the traverse the link. Since the data can t be sent immediately.11g link may use 54 Mbps radios. Note that web email services (such as Yahoo or Hotmail) should be considered as web browsing. or simply bandwidth (although this term is unrelated to radio bandwidth). IM will tolerate high latency. As with IM. How much throughput will your users really need? It depends on how many users you have. As web browsers request more data (large images. It is important to understand that a wireless device s listed speed (the data rate) refers to the rate at which the radios can exchange symbols. Communication is asynchronous. and high latency is commonly referred to as lag. Throughput is also referred to as channel capacity. Application Text messaging / IM BW / User < 1 kbps Notes As traffic is infrequent and asynchronous. so it will tolerate latency. viruses. Web browsers only use the network when data is requested. Your link will eventually send all of the queued traffic. but your users will likely complain as the lag increases. Email 1 to 100 kbps Web browsing 50 .11g protocol. dialup. The time that it takes for data to traverse a link is called latency.) bandwidth usage will go up significantly. but it will only provide up to 22 Mbps of actual throughput.100+ kbps . The rest is overhead that the radios need in order to coordinate their signals using the 802. and transmitted as quickly as possible. and spam significantly add to bandwidth usage. Note that throughput is a measurement of bits over time. not the usable throughput you will observe.

as quickly as possible.5 to 5 Mbps or more of throughput at peak times. some intermittent latency is avoided by using buffers on the client. Lag greater than a few milliseconds is unacceptable for VoIP. never both at once) you should accordingly double the required throughput. Your wireless links must provide that capacity every second. multiply the expected number of users by the sort of application they will probably use. Since all of your users are unlikely to use the connection at precisely the same moment. Latency on a VoIP connection is immediate and annoying to users. KaZaA. it only transmits or receives. But extended periods of lag will cause audio “skips” or outright session failures. it is common practice to oversubscribe available throughput by some factor (that is. While peer to peer applications will tolerate any amount of latency. the bandwidth is used roughly equally in both directions. or conversations will lag.infinite Mbps To estimate the necessary throughput you will need for your network. Since 802. they tend to use up all available throughput by transmitting data to as many clients as possible. for a total of 10 Mbps. For example. Gnutella. eDonkey. It can tolerate some transient latency by using large buffers on the client.160 kbps Notes Each user of a streaming audio service will use a constant amount of relatively large bandwidth for as long as it plays.) 0 . 50 users who are chiefly browsing the web will likely consume 2. As with streaming audio. On the other hand.66 Chapter 3: Network Design Application Streaming audio BW / User 96 . As with streaming audio. Streaming video requires high throughput and low latency to work properly. etc. Use of these applications will cause latency and throughput problems for all other network users unless you use careful bandwidth shaping. and will tolerate some latency.11g wireless equipment is half duplex (that is.100+ kbps Streaming video 64 . allow more users than the maximum available band- .200+ kbps Peer-to-peer filesharing applications (BitTorrent. VoIP commits a constant amount of bandwidth to each user for the duration of the call. 50 simultaneous VoIP users would require 5 Mbps or more of throughput in both directions with absolutely no latency. Voice over IP (VoIP) 24 . But with VoIP.

the ICTP has put together a bandwidth simulator. your users will eventually find applications that will use it all. you can significantly reduce latency and increase overall network throughput. To get a feeling for the lag felt on very slow connections. called path loss. Online databases such as the one provided by SeattleWireless (http://www. but can sometimes be difficult to find.seattlewireless. By carefully monitoring throughput throughout your network. This demonstration gives you an immediate understanding of how low throughput and high latency reduce the usefulness of the Internet as a communications tool. each with its associated antenna. Antennas have the same characteristics when receiving and transmitting. Whether or not signals can be passed between the radios depends on the quality of the equipment being used and on the diminishment of the signal due to distance. Oversubscribing by a factor of 2 to 5 is quite common. It is available at http://wireless. It is expressed in milliwatts or in dBm. • Antenna Gain. It will simultaneously download a web page at full speed and at a reduced rate that you choose. and other techniques.ictp. web caching. By using bandwidth shaping. you will oversubscribe by some amount when building your network infrastructure. Expect that no matter how much capacity you supply. the two being separated by the path to be covered.Chapter 3: Network Design 67 width can support).net/HardwareComparison) may help. the radios require a certain minimum signal to be collected by the antennas and presented to their input socket. and how much additional resources will be needed. The TX power of a given device should be specified in the literature provided by the manufacturer. In order to have a communication between the two. Determining if the link is feasible is a process called link budget calculation. In all likelihood. you will be able to plan when to upgrade various parts of the network. Transmit Power ranges from 30mW to 200mW or more. TX power is often dependent on the transmission rate. So a 12 dBi antenna is simply . As we ll see at the end of this chapter. Calculating the link budget The power available in an 802.trieste. Antennas are passive devices that create the effect of amplification by virtue of their physical shape.it/simulator/ Link planning A basic communication system consists of two radios.11 system can be characterized by the following factors: • Transmit Power. using bandwidth shaping techniques can help mitigate some latency problems.

The most convenient way to express its contribution to the total loss is by adding an “allowed loss” to the free space. The minimum RSL is always expressed as a negative dBm (. . windows and floors of buildings. • Minimum Received Signal Level.68 Chapter 3: Network Design a 12 dBi antenna. or simply. the sensitivity of the receiver. It is better to have cables as short as possible. and as a general rule the lowest rate (1 Mbps) has the greatest sensitivity. Like TX power. This loss happens because the radiated signal energy expands as a function of the distance from the transmitter. Parabolic antennas have a gain of 19-24 dBi. When calculating the path loss. without specifying if it is in transmission or reception mode. in the range of 2-3 dB. while walls contribute 10 to 15 dB depending upon the construction. The second contribution to the path loss is given by attenuation. Ignoring everything else. commonly known as free space loss. the RSL specifications should be provided by the manufacturer of the equipment. Some of the RF energy reaches the receiving antenna directly. going from the radios to the antennas. depending only on the distance. Some of the signal s energy is lost in the cables. the equation for the free space loss is Lfsl = 40 + 20*log(r) where Lfsl is expressed in dB and r is the distance between the transmitter and receiver. omnidirectional antennas have 5-12 dBi. Signal loss for short coaxial cables including connectors is quite low. sectorial antennas have roughly a 12-15 dBi gain. The loss depends on the type of cable used and on its length. This is independent from the environment. Signal power is diminished by geometric spreading of the wavefront. For example. several effects must be considered.dBm) and is the lowest power of signal the radio can distinguish.45 GHz as the signal frequency. the smaller the received signal is due to free space loss. the connectors and other devices. One has to take into account the free space loss. the further away the two radios. the RF energy leaves the transmitting antenna and energy spreads out. The minimum will be typically in the range of -75 to -95 dBm. This takes place as some of the signal power is absorbed when the wave passes through solid objects such as trees. walls. Along the link path. in meters. Attenuation can vary greatly depending upon the structure of the object the signal is passing through. The minimum RSL is dependent upon rate. Using decibels to express the loss and using 2. and it is very difficult to quantify. attenuation and scattering. experience shows that trees add 10 to 20 dB of loss per tree in the direct path. • Cable Losses.

If you are using different radios on either side of the link. even if the other is receiving a distorted one. Multipath is in fact a very location-specific phenomenon. you should calculate the path loss twice. The distortion given by multipath degrades the ability of the receiver to recover the signal in a manner much like signal loss. and scattering are combined. In commercial devices. Since the reflected signal has a longer way to travel. attenuation. An exponent of 3 can be used in an outdoor environment with trees. If there are two antennas. To evaluate if a link is feasible. or signal dispersion. the signal at the receiving antenna can be zeroed by the reflected signals. antenna switching diversity is used: there are multiple antennas on multiple inputs. they will not add destructively at a second. This effect is called multipath. the path loss is: L(dB) = 40 + 10*n*log(r) + L(allowed) For a rough estimate of the link feasibility. In some cases. the radio uses the antenna last used for reception. If two signals add out of phase at one location. Note that when performing this calculation. at least one of them should be able to receive a usable signal. Adding up all the gains and subtracting all the losses gives + + TX Power Radio 1 Antenna Gain Radio Cable Losses Radio Antenna Gain Radio Cable Losses Radio = Total Gain 1 1 2 2 . It consists of adding a second antenna to the radio. one must know the characteristics of the equipment being used and evaluate the path loss. you should only add the TX power of one side of the link. When free space loss. called antenna diversity. one can evaluate just the free space loss. When they add together out of phase. Part of the RF energy which bounces off the ground reaches the receiving antenna. The environment is in fact a very important factor.Chapter 3: Network Design 69 while some bounces off the ground. The exponent tends to increase with the range in an environment with a lot of scattering. while one of 4 can be used for an indoor environment. A simple way of applying the effects of scattering in the calculation of the path loss is to change the exponent of the distance factor of the free space loss formula. The environment can bring further signal loss. This is known as extreme fading. the received signal is almost worthless. and should be considered for an exact evaluation of the link. once for each direction (using the appropriate TX power for each calculation). There is a simple technique that is used to deal with multipath. with a single receiver. The signal is thus received through only one antenna at a time. In some cases reflected signals add together and cause no problem. When transmitting. it arrives at the receiving antenna later than the direct signal. nearby location. and should never be neglected. or nulling.

then the link is feasible! The received signal is powerful enough for the radios to use it. The access point is connected to an omnidirectional antenna with 10 dBi gain. a margin of 20dB should be safe enough. The transmitting power of the AP is 100mW (or 20 dBm) and its sensitivity is -89 dBm. while the client is connected to a sectorial antenna with 14 dBi gain. This margin is the amount of signal above the sensitivity of radio that should be received in order to ensure a stable. Example link budget calculation As an example. The transmitting power of the client is 30mW (or 15 dBm) and its sensitivity is -82 dBm. with a loss of 2dB at each side. Substitute the transmit power for that of the second radio. A margin of 10 to 15 dB is fine. considering only the free space loss is: . with one access point and one client radio. On a given path.Path Loss = Signal Level at one side of the link If the resulting signal level is greater than the minimum received signal level. so -56 dBm is greater than -70 dBm. The cables are short. high quality radio link during bad weather and other atmospheric disturbances. so a certain margin (difference between the signal level and the minimum received signal level) should be considered.2 dB (TX Power Radio 1) (Antenna Gain Radio (Cable Losses Radio (Antenna Gain Radio (Cable Losses Radio 1) 1) 2) 2) 40 dB = Total Gain The path loss for a 5 km link. the variation in path loss over a period of time can be large. Remember that the minimum RSL is always expressed as a negative dBm.2 dB + 14 dBi . we want to estimate the feasibility of a 5 km link. and compare the result against the minimum received signal level of the first radio. repeat the calculation for the other direction.70 Chapter 3: Network Design Subtracting the Path Loss from the Total Gain: Total Gain . Once you have calculated the link budget in one direction. Adding up all the gains and subtracting all the losses for the AP to client link gives: 20 dBm + 10 dBi . To give some space for attenuation and multipath in the received radio signal.

Next we calculate the link from the client back to the access point: 15 dBm + 14 dBi .terabeam.113 dB = -78 dB Since the receive sensitivity of the AP is -89dBm.cgi) is an excellent tool.main.2 dB + 10 dBi .78dB). you will get an additional 10dBi of gain on both directions of the link (remember.athenet. but note that adding an amplifier or higher powered card to one end generally does not help the overall quality of the link. The calculation scripts can even be downloaded from the website and installed locally. this link will probably work but could use a bit more gain. the signal level is just enough for the client radio to be able to hear the access point. the Green Bay Professional Packet Radio s Wireless Network Link Analysis (http://my. For example.php). The Terabeam website also has excellent calculators available online (http://www. There is only 9 dB of margin (82 dB . . the path loss is the same on the return trip. The Super Edition generates a PDF file containing the Fresnel zone and radio path graphs. antenna gain is reciprocal).113 dB = -73 dB 71 Since -73 dB is greater than the minimum receive sensitivity of the client radio (-82 dBm). By using a 24dBi dish on the client side rather than a 14dBi sectorial antenna. but may not be enough to protect against extreme weather conditions.73 dB) which will likely work fine in fair weather. Overall. this leaves us 11dB of fade margin (89dB .Chapter 3: Network Design Path Loss = 40 + 20log(5000) = 113 dB Subtracting the path loss from the total gain 40 dB .com/support/calculations/index.net/~multiplx/cgi-bin/wireless. Online tools can be used to calculate the link budget.2 dB (TX Power Radio 2) (Antenna Gain Radio (Cable Losses Radio (Antenna Gain Radio (Cable Losses Radio 2) 2) 1) 1) 35 dB = Total Gain Obviously. So our received signal level on the access point side is: 35 dB . A more expensive option would be to use higher power radios on both ends of the link.

000 3. see Appendix C.000 80 94 100 110 113 120 For more path loss distances.Total Loss = Signal > Radio 2 Sensitivity . simply approximate your link distance.72 Chapter 3: Network Design Tables for calculating link budget To calculate the link budget. then fill in the following tables: Free Space Path Loss at 2. Antenna Gain: Radio 1 Antenna + Radio 2 Antenna = Total Antenna Gain Losses: Radio 1 + Cable Loss (dB) Radio 2 + Cable Loss (dB) Free Space Path Loss (dB) = Total Loss (dB) Link Budget for Radio 1 Radio 1 TX Power + Antenna Gain Radio 2: .4 GHz Distance (m) Loss (dB) 100 500 1.000 10.000 5.

qsl. these tools will take many other relevant factors into account as well (such as tree absorption. and RadioMobile. To begin. You can browse these tools online at http://www.athenet. Wireless Network Link Analysis. while the tool asks for the highest transmitted frequency. including the transmission line type.cgi. there are a number of tools available that will help automate the process. Next. This data will be used for calculating the antenna tilt . enter the channel to be used on the link. This can be specified in MHz or GHz.Chapter 3: Network Design 73 Link Budget for Radio 2 Radio 2 TX Power + Antenna Gain Radio 1: .Total Loss = Signal > Radio 1 Sensitivity If the received signal is greater than the minimum received signal strength in both directions of the link. so feel free to use the center frequency instead. Since the tools are available online. To find the highest transmitted frequency for a channel. we will discuss two free tools that are useful for planning wireless links: Green Bay Professional Packet Radio s online interactive network design utilities. they will work with any device that has a web browser and Internet access. terrain effects. In this section. just add 11MHz to the center frequency. In addition to calculating free space loss. climate. and even estimating path loss in urban areas).net/n9zia/wireless/page09. antenna gain. You can also enter the antenna height and elevation for this site. Note that the table lists the channel s center frequency.main. consult the table in Appendix B. and other details. then the link is possible. Interactive design CGIs The Green Bay Professional Packet Radio group (GBPRR) has made a variety of very useful link planning tools available for free online. The difference in the ultimate result is minimal.html . as well as any noise received along the path. enter the details for the transmitter side of the link. We will look at the first tool. in detail. Link planning software While calculating a link budget by hand is straightforward. You can find it online at http://my.net/~multiplx/cgi-bin/wireless. Try to fill in as much data as you know or can estimate. If you don t know the frequency.

During the simulation. including net master/ slave. Source code to most of the tools is provided as well. In addition to the basic link analysis tool. Finally. click the Submit button for a detailed report about the proposed link. error rates. For calculating Fresnel zone clearance. and point-to-multipoint. GBPRR provides a “super edition” that will produce a PDF report. indicating received signal strength at various points along the path. Enter as much data as you know or can estimate. or entered by hand. you can play “what-if?” to see how changing various parameters will affect the connection. It predicts the performance of a radio link by using information about the equipment and a digital map of the area. including losses due to obstacles. and distance of the link. DEMs do not show coastlines or other readily identifiable landmarks. It automatically builds a profile between two points in the digital map showing the coverage area and first Fresnel zone. Enter all available data in the appropriate fields. or using Linux and the Wine emulator. terrain. It is public domain software that runs on Windows. It works for systems having frequencies from 100 kHz to 200 GHz. RadioMobile Radio Mobile is a tool for the design and simulation of wireless systems. and uptime. The software calculates the coverage area from the base station in a point-to-multipoint system. By adjusting values on the form. The next section is very similar.74 Chapter 3: Network Design angle. It is possible to create networks of different topologies. you will need to use GBPRR s Fresnel Zone Calculator. Radio Mobile uses a digital terrain elevation model for the calculation of coverage. but they can easily be combined with other kinds of data (such as aerial photos or topographical charts) in several layers to obtain a more useful and readily recognizable representation. as well as a number of other very useful tools (including the Fresnel Zone Calculator. as well as the projected path loss. but will give you a rough idea of the feasibility of the link. Distance & Bearing Calculator. the last section describes the climate. This includes all of the data entered. point-to-point. These numbers are all completely theoretical. Digital elevation maps (DEM) are available for free from several sources. Link distance can be calculated by specifying the latitude and longitude of both sites. but includes information about the other end of the link. it checks for line of sight and calculates the Path Loss. and are available for most of the world. and Decibel Conversion Calculator to name just a few). The digital elevation maps can be merged with . You can digitize your own maps and combine them with DEMs. Now.

10).DLL file from a Windows machine that already has the Visual Basic 6 run-time environment installed. satellite photos and Internet map services (such as Google Maps) to produce accurate prediction plots. from the Ubuntu Universe repository There are detailed instructions for installing RadioMobile on Windows at http://www. with examples and tutorials. these are extra steps only needed for Windows users.org/rmw/english1.com/ • Wine version 20050725. is available at: http://www.cplus.org/rmw/english1.html. using RadioMobile. You will either need to copy the MSVBVM60. making sure to unzip the downloaded files in the same directory into which you have placed the downloaded DLL file.cplus. Note that you don't have to worry about the stuff after step 4.DLL. The main Radio Mobile webpage.ubuntu. Figure 3. You should follow all of the steps except for step 1 (since it is difficult to extract a DLL from the VBRUN60SP6. http://www.20: Link feasibility. We were able to make Radio Mobile work with Linux using the following environment: • IBM Thinkpad x31 • Ubuntu Breezy (v5. . Now continue with step 2 at from the above URL. or simply Google for MSVBVM60.EXE file under Linux). some button labels may run beyond the frame of the button and can be hard to read.Chapter 3: Network Design 75 scanned maps.html RadioMobile under Linux Radio Mobile will also work using Wine under Ubuntu Linux. including Fresnel zone and line of sight estimate. and download the file. While the application runs.

6 An omnidirectional antenna receives noise from all directions Multiple sectorial antennas help to mitigate noise and add additional bandwidth Figure 3. you can reduce the overall noise received at a distribution point. By making use of several sectorial antennas.4 GHz band. many consumer devices use it for a wide range of applications. 6 ch . Using omnidirectional antennas will receive noise from all directions. Bluetooth. • Increase antenna gain on both sides of a point-to-point link. can cause significant problems for long range wireless links. These signals. Here are some steps you can use to reduce reception of unwanted signals. • Use sectorials instead of using an omnidirectional. you can start Wine from a terminal with the command: # wine RMWDLX.1 ch ch . Two high gain dishes that are pointed at each other will reject noise from directions that are outside the path of the link. analog video senders.76 Chapter 3: Network Design Finally. 1 .1 ch . multiple sectorials.1 1 .1 ch. as well as other local wireless networks. and even microwave ovens compete with wireless data networks for use of the very limited 2. but their increased directionality tends to reject noise from areas around the link. Antennas not only add gain to a link.exe You should see RadioMobile running happily in your XWindows session. By staggering the channels used on each sectorial. baby monitors. ch.21: A single omnidirectional antenna vs. Avoiding noise The unlicensed ISM and U-NII bands represent a very tiny piece of the known electromagnetic spectrum. Since this region can be utilized without paying license fees. Cordless phones. you can also increase the available bandwidth to your clients.

trees. such as Ronja (http://ronja. Remember that 802. We will see some examples of these tools in Chapter 6. While it may be possible to create a 12 km link that cuts across the middle of a city. If you can find it.4 GHz. If you can break that link into two or three shorter hops. use licensed spectrum. you will likely have all kinds of interference problems. Of course. For long distance point-to-point links that require very high throughput and maximum uptime. or other obstacles in the path of a long distance link. In these cases. there is currently far more consumer equipment installed in the field that uses 2.twibright.8 GHz.Chapter 3: Network Design 77 • Don t use an amplifier. you need tools that will show you what is happening in the air at 2. but are only separated by 5MHz. Other technologies. • Use the best available channel. . Remember that the wireless landscape can change at any time as people add new devices (cordless phones. or another unlicensed band. Terrestrial microwave systems simply cannot tolerate large hills. • If possible. use 5. these features come at a much higher price tag compared to unlicensed equipment. some old 802. Perform a site survey.4 GHz.11b/g channels are 22 MHz wide. this is certainly an option.11 equipment uses unlicensed spectrum at 900MHz (unfortunately at much lower bit rates). • Use smaller hops and repeaters. While this is only a short term solution. including sources of interference. There are places where all available unlicensed spectrum is effectively used.8 GHz step-up device will let you avoid this congestion altogether. 900MHz. Repeaters The most critical component to building long distance network links is line of sight (often abbreviated as LOS). To identify sources of noise. and select a channel that is as far as possible from existing sources of interference.11a or a 2. but noise problems are also unlikely in those settings. it may make sense to spend the additional money for proprietary equipment that uses a less congested band. As we will see in Chapter 4. etc.com/) use optical technology for short distance. Keep your point-to-point links as short as possible. Amplifiers also cause interference problems for other nearby users of the band. other networks. you may need to perform another site survey and pick a different channel. You must have a clear idea of the lay of the land between two points before you can determine if a link is even possible. rather than a single long distance shot.4 GHz to 5. the link will likely be more stable. • If all else fails. noise-free links. Using 802. Obviously this isn t possible on long distance rural links where power and mounting structures are unavailable.) If your link suddenly has trouble sending packets. amplifiers can make interference issues worse by indiscriminately amplifying all received signals.

Arrangements can often be worked out with building owners to provide bandwidth in exchange for roof rights and electricity. If you can t go over or through an obstacle. But depending on your network layout. but never both at once. as long as they are each configured to use nonoverlapping channels. In a mesh network. try a multi-hop approach to avoid the obstacle. In a traditional infrastructure network. Rather than using a direct link. Repeaters can be purchased as a complete hardware solution. . For example. These devices are cheaper. When using a single radio (called a one-arm repeater). or easily assembled by connecting two or more wireless nodes together with Ethernet cable. to allow multiple clients to connect to either side of the repeater. one or more devices may need to use ad-hoc or even client mode. repeaters are used to overcome obstacles in the path of a long distance link. If the building owner isn t interested. Repeaters are nodes that are configured to rebroadcast traffic that is not destined for the node itself. Mountains may block your signal. but those buildings contain people. tenants on high floors may be able to be persuaded to install equipment in a window. remember that nodes must be configured for master. remember that obstacles can sometimes be turned into assets. Typically. simpler. A repeater with two (or more) radio cards can operate all radios at full capacity. repeaters can also supply an Ethernet connection to provide local connectivity. Of course. nodes must be configured to pass along traffic to other nodes. overall efficiency is slightly less than half of the available bandwidth. and have lower power requirements. there may be buildings in your path. since the radio can either send or receive data.11 technology. Repeater Figure 3. you can often go around it. both radios in a repeater are configured for master mode.22: The repeater forwards packets over the air between nodes that have no direct line of sight. but assuming power can be provided they also make very good repeater sites.78 Chapter 3: Network Design But even if there is a mountain between two points. When planning to use a repeater built with 802. A repeater can use one or more wireless devices. managed. or ad-hoc mode. Typically. every node is a repeater.

bandwidth available on any link approaches infinity. Traffic optimization Bandwidth is measured as the amount of bits transmitted over a time interval. B. nodes A. you need only wait long enough. Finally.23: No power was available at the top of the hill. and as long as your node is in range of the repeater. and D can communicate with each other. Note that traffic from node D actually travels further away from the rest of the network before the repeater forwards it along. the bandwidth provided by any given network connection is not infinite.Chapter 3: Network Design 79 Repeater Repeater Figure 3. human users are not as patient as computers. you need only contact five intermediaries before finding the person. Of course. you can communicate with any node the repeater can reach. for any given period of time. since site C is in the way and is not hosting a node. This means that over time. Repeaters in high places can “see” a great deal of intermediaries. you may need to consider going backwards in order to go forwards. This idea says that no matter who you are looking for. By installing a high repeater. Repeaters in networks remind me of the “six degrees of separation” principle. You can always download (or upload) as much traffic as you like. If there is a high site available in a different direction. Repeater C A B D Figure 3. a stable link can be made via an indirect route. and are not willing to . and that site can see beyond the obstacle.24: Site D could not make a clean link to site A or B. but it was circumvented by using multiple repeater sites around the base. Unfortunately.

It is usually the task of the network administrator to deal with user perception issues like these. or other dynamically generated content. or Webalizer. adding content filtering and advertisement blocking). It is free.for example. Web caching A web proxy server is a server on the local network that keeps copies of recently retrieved or often used web pages. first showing some text and then displaying the graphics one by one. a typical page begins to load slowly. bandwidth must be managed and prioritized much like any other limited resource. reliable. This happens because the information is sent to the computer so quickly that it spends a perceptible amount of time rendering the page. But unless this is explained to some impatient users.80 Chapter 3: Network Design wait an infinite amount of time for their information to traverse the network. For a more thorough discussion of the complex subject of bandwidth optimization. commonly used software packages: These are the most • Squid. When the next person retrieves these pages. The apparent loading of web pages is also affected. Squid produces logs that can be analyzed using software such as Awstats. Proxy server products There are a number of web proxy servers available. both of which are open source and produce good graphical reports. Open source Squid is the de facto standard at universities. there could be a delay when nothing seems to happen. it is easier to install as part of the distribution than to download it from . When a proxy server is implemented. The overall time it takes to load the whole page might take only ten seconds (whereas without a proxy server. With a slow Internet link. pages that are the output of serverside scripts. easy to use and can be enhanced (for example. In most cases. For this reason. You will significantly improve response time and maximize available throughput by eliminating unwanted and redundant traffic from your network. This results in significantly faster web access in most cases. they may say the proxy server has made things slower.net/). while reducing overall Internet bandwidth usage. they are served from the local proxy server instead of from the Internet. and then the page will load almost at once. or parts of pages. This section describes a few common techniques for making sure that your network carries only the traffic that must traverse it. In a network with a proxy server. it may take 30 seconds to load the page gradually). the administrator should also be aware that some pages are not cacheable-. see the free book How to Accelerate Your Internet (http://bwmo.

Chapter 3: Network Design

81

http://www.squid-cache.org/ (most Linux distributions such as Debian, as well as other versions of Unix such as NetBSD and FreeBSD come with Squid). A good Squid configuration guide can be found on the Squid Users Guide Wiki at http://www.deckle.co.za/squid-users-guide/. • Microsoft Proxy server 2.0. Not available for new installations because it has been superseded by Microsoft ISA server and is no longer supported. It is nonetheless used by some institutions, although it should perhaps not be considered for new installations. • Microsoft ISA server. ISA server is a very good proxy server program, that is arguably too expensive for what it does. However, with academic discounts it may be affordable to some institutions. It produces its own graphical reports, but its log files can also be analyzed with popular analyzer software such as Sawmill (http://www.sawmill.net/). Administrators at a site with MS ISA Server should spend sufficient time getting the configuration right; otherwise MS ISA Server can itself be a considerable bandwidth user. For example, a default installation can easily consume more bandwidth than the site has used before, because popular pages with short expiry dates (such as news sites) are continually being refreshed. Therefore it is important to get the pre-fetching settings right, and to configure pre-fetching to take place mainly overnight. ISA Server can also be tied to content filtering products such as WebSense. For more information, see: http://www.microsoft.com/isaserver/ and http://www.isaserver.org/ .

Preventing users from bypassing the proxy server
While circumventing Internet censorship and restrictive information access policy may be a laudable political effort, proxies and firewalls are necessary tools in areas with extremely limited bandwidth. Without them, the stability and usability of the network are threatened by legitimate users themselves. Te c h n i q u e s f o r b y p a s s i n g a p r o x y s e r v e r c a n b e f o u n d a t http://www.antiproxy.com/ . This site is useful for administrators to see how their network measures up against these techniques. To enforce use of the caching proxy, you might consider simply setting up a network access policy and trusting your users. In the layout below, the administrator has to trust that his users will not bypass the proxy server. In this case the administrator typically uses one of the following techniques: • Not giving out the default gateway address through DCHP. This may work for a while, but some network-savvy users who want to bypass the proxy might find or guess the default gateway address. Once that happens, word tends to spread about how to bypass the proxy.

82

Chapter 3: Network Design

• Using domain or group policies. This is very useful for configuring the correct proxy server settings for Internet Explorer on all computers in the domain, but is not very useful for preventing the proxy from being bypassed, because it depends on a user logging on to the NT domain. A user with a Windows 95/98/ME computer can cancel his log-on and then bypass the proxy, and someone who knows a local user password on his Windows NT/2000/XP computer can log on locally and do the same. • Begging and fighting with users. This approach, while common, is never an optimal situation for a network administrator.

Internet

Router

PC

PC

PC

Proxy Server

Figure 3.25: This network relies on trusted users to properly configure their PCs to use the proxy server.

The only way to ensure that proxies cannot be bypassed is by using the correct network layout, by using one of the three techniques described below.

Firewall
A more reliable way to ensure that PCs don t bypass the proxy can be implemented using the firewall. The firewall can be configured to allow only the proxy server to make HTTP requests to the Internet. All other PCs are blocked, as shown in Figure 3.26. Relying on a firewall may or may not be sufficient, depending on how the firewall is configured. If it only blocks access from the campus LAN to port 80 on web servers, there will be ways for clever users to find ways around it. Additionally, they will be able to use other bandwidth hungry protocols such as BitTorrent or Kazaa.

Chapter 3: Network Design

83

Internet Proxy Server Proxy server is granted full access

Direct access is forbidden by the firewall

Firewall

x

PC

PC

PC

Figure 3.26: The firewall prevents PCs from accessing the Internet directly, but allows access via the proxy server.

Two network cards
Perhaps the most reliable method is to install two network cards in the proxy server and connect the campus network to the Internet as shown below. In this way, the network layout makes it physically impossible to reach the Internet without going through the proxy server.

Internet

Proxy server

Figure 3.27: The only route to the Internet is through the proxy.

84

Chapter 3: Network Design

The proxy server in this diagram should not have IP forwarding enabled, unless the administrators knows exactly what they want to let through. One big advantage to this design is that a technique known as transparent proxying can be used. Using a transparent proxy means that users web requests are automatically forwarded to the proxy server, without any need to manually configure web browsers to use it. This effectively forces all web traffic to be cached, eliminates many chances for user error, and will even work with devices that do not support use of a manual proxy. For more details about configuring a transparent proxy with Squid, see: • http://www.squid-cache.org/Doc/FAQ/FAQ-17.html • http://tldp.org/HOWTO/TransparentProxy.html

Policy-based routing
One way to prevent bypassing of the proxy using Cisco equipment is with policy routing. The Cisco router transparently directs web requests to the proxy server. This technique is used at Makerere University. The advantage of this method is that, if the proxy server is down, the policy routes can be temporarily removed, allowing clients to connect directly to the Internet.

Mirroring a website
With permission of the owner or web master of a site, the whole site can be mirrored to a local server overnight, if it is not too large. This is something that might be considered for important websites that are of particular interest to the organization or that are very popular with web users. This may have some use, but it has some potential pitfalls. For example, if the site that is mirrored contains CGI scripts or other dynamic content that require interactive input from the user, this would cause problems. An example is a website that requires people to register online for a conference. If someone registers online on a mirrored server (and the mirrored script works), the organizers of the site will not have the information that the person registered. Because mirroring a site may infringe copyright, this technique should only be used with permission of the site concerned. If the site runs rsync, the site could be mirrored using rsync. This is likely the fastest and most efficient way to keep site contents synchronized. If the remote web server is not running rsync, the recommended software to use is a program called wget. It is part of most versions of Unix/Linux. A Windows version can be found at http://xoomer.virgilio.it/hherold/, or in the free Cygwin Unix tools package (http://www.cygwin.com/).

Chapter 3: Network Design

85

A script can be set up to run every night on a local web server and do the following: • Change directory to the web server document root: for example, /var/ www/ on Unix, or C:\Inetpub\wwwroot on Windows. • Mirror the website using the command:
wget --cache=off -m http://www.python.org

The mirrored website will be in a directory www.python.org. The web server should now be configured to serve the contents of that directory as a name-based virtual host. Set up the local DNS server to fake an entry for this site. For this to work, client PCs should be configured to use the local DNS server(s) as the primary DNS. (This is advisable in any case, because a local caching DNS server speeds up web response times).

Pre-populate the cache using wget
Instead of setting up a mirrored website as described in the previous section, a better approach is to populate the proxy cache using an automated process. This method has been described by J. J. Eksteen and J. P. L. Cloete of the CSIR in Pretoria, South Africa, in a paper entitled Enhancing International World Wide Web Access in Mozambique Through the Use of Mirroring and Caching Proxies. In this paper (available at http://www.isoc.org/inet97/ans97/cloet.htm) they describe how the process works: "An automatic process retrieves the site's home page and a specified number of extra pages (by recursively following HTML links on the retrieved pages) through the use of a proxy. Instead of writing the retrieved pages onto the local disk, the mirror process discards the retrieved pages. This is done in order to conserve system resources as well as to avoid possible copyright conflicts. By using the proxy as intermediary, the retrieved pages are guaranteed to be in the cache of the proxy as if a client accessed that page. When a client accesses the retrieved page, it is served from the cache and not over the congested international link. This process can be run in off-peak times in order to maximize bandwidth utilization and not to compete with other access activities." The following command (scheduled to run at night once every day or week) is all that is needed (repeated for every site that needs pre-populating).
wget --proxy-on --cache=off --delete after -m http://www.python.org

These options enable the following:

86

Chapter 3: Network Design

• -m: Mirrors the entire site. wget starts at www.python.org and follows all hyperlinks, so it downloads all subpages. • --proxy-on: Ensures that wget makes use of the proxy server. This might not be needed in set-ups where a transparent proxy is employed. • --cache=off: Ensures that fresh content is retrieved from the Internet, and not from the local proxy server. • --delete after: Deletes the mirrored copy. The mirrored content remains in the proxy cache if there is sufficient disk space, and the proxy server caching parameters are set up correctly. In addition, wget has many other options; for example, to supply a password for websites that require them. When using this tool, Squid should be configured with sufficient disk space to contain all the pre-populated sites and more (for normal Squid usage involving pages other than the pre-populated ones). Fortunately, disk space is becoming ever cheaper and disk sizes are far larger than ever before. However, this technique can only be used with a few selected sites. These sites should not be too big for the process to finish before the working day starts, and an eye should be kept on disk space.

Cache hierarchies
When an organization has more than one proxy server, the proxies can share cached information among them. For example, if a web page exists in server A's cache, but not in the cache of server B, a user connected via server B might get the cached object from server A via server B. Inter-Cache Protocol (ICP) and Cache Array Routing Protocol (CARP) can share cache information. CARP is considered the better protocol. Squid supports both protocols, and MS ISA Server supports CARP. For more information, see http://squid-docs.sourceforge.net/latest/html/c2075.html. This sharing of cached information reduces bandwidth usage in organizations where more than one proxy is used.

Proxy specifications
On a university campus network, there should be more than one proxy server, both for performance and also for redundancy reasons. With today's cheaper and larger disks, powerful proxy servers can be built, with 50 GB or more disk space allocated to the cache. Disk performance is important, therefore the fastest SCSI disks would perform best (although an IDE based cache is better than none at all). RAID or mirroring is not recommended. It is also recommended that a separate disk be dedicated to the cache. For example, one disk could be for the cache, and a second for the operating system and cache logging. Squid is designed to use as much RAM as it can get, because when data is retrieved from RAM it is much faster than when it

Chapter 3: Network Design

87

comes from the hard disk. For a campus network, RAM memory should be 1GB or more: • Apart from the memory required for the operating system and other applications, Squid requires 10 MB of RAM for every 1 GB of disk cache. Therefore, if there is 50 GB of disk space allocated to caching, Squid will require 500 MB extra memory. • The machine would also require 128 MB for Linux and 128 MB for Xwindows. • Another 256 MB should be added for other applications and in order that everything can run easily. Nothing increases a machine's performance as much as installing a large amount of memory, because this reduces the need to use the hard disk. Memory is thousands of times faster than a hard disk. Modern operating systems keep frequently accessed data in memory if there is enough RAM available. But they use the page file as an extra memory area when they don't have enough RAM.

DNS caching and optimization
Caching-only DNS servers are not authoritative for any domains, but rather just cache results from queries asked of them by clients. Just like a proxy server that caches popular web pages for a certain time, DNS addresses are cached until their time to live (TTL) expires. This will reduce the amount of DNS traffic on your Internet connection, as the DNS cache may be able to satisfy many of the queries locally. Of course, client computers must be configured to use the caching-only name server as their DNS server. When all clients use this server as their primary DNS server, it will quickly populate a cache of IP addresses to names, so that previously requested names can quickly be resolved. DNS servers that are authoritative for a domain also act as cache name-address mappings of hosts resolved by them.

Bind (named)
Bind is the de facto standard program used for name service on the Internet. When Bind is installed and running, it will act as a caching server (no further configuration is necessary). Bind can be installed from a package such as a Debian package or an RPM. Installing from a package is usually the easiest method. In Debian, type
apt-get install bind9

In addition to running a cache, Bind can also host authoritative zones, act as a slave to authoritative zones, implement split horizon, and just about everything else that is possible with DNS.

88

Chapter 3: Network Design

dnsmasq
One alternative caching DNS server is dnsmasq. It is available for BSD and most Linux distributions, or from http://www.thekelleys.org.uk/dnsmasq/. The big advantage of dnsmasq is flexibility: it easily acts as both a caching DNS proxy and an authoritative source for hosts and domains, without complicated zone file configuration. Updates can be made to zone data without even restarting the service. It can also serve as a DHCP server, and will integrate DNS service with DHCP host requests. It is very lightweight, stable, and extremely flexible. Bind is likely a better choice for very large networks (more than a couple of hundred nodes), but the simplicity and flexibility of dnsmasq makes it attractive for small to medium sized networks.

Windows NT
To install the DNS service on Windows NT4: select Control Panel Network Services Add Microsoft DNS server. Insert the Windows NT4 CD when prompted. Configuring a caching-only server in NT is described in Knowledge Base article 167234. From the article: "Simply install DNS and run the Domain Name System Manager. Click on DNS in the menu, select New Server, and type in the IP address of your computer where you have installed DNS. You now have a cachingonly DNS server."

Windows 2000
Install DNS service: Start Settings Control Panel Add/Remove Software. In Add/Remove Windows Components, select Components Networking Services Details Domain Name System (DNS). Then start the DNS MMC (Start Programs Administrative Tools DNS) From the Action menu select "Connect To Computer..." In the Select Target Computer window, enable "The following computer:" and enter the name of a DNS server you want to cache. If there is a . [dot] in the DNS manager (this appears by default), this means that the DNS server thinks it is the root DNS server of the Internet. It is certainly not. Delete the . [dot] for anything to work.

Split DNS and a mirrored server
The aim of split DNS (also known as split horizon) is to present a different view of your domain to the inside and outside worlds. There is more than one way to do split DNS; but for security reasons, it's recommended that you have two separate internal and external content DNS servers (each with different databases). Split DNS can enable clients from a campus network to resolve IP addresses for the campus domain to local RFC1918 IP addresses, while the rest of the

ac. The DNS records for this external zone would look like this: makerere.ug www CNAME ftp CNAME mail CNAME mailserver webserver ftpserver webserver. host websites on behalf of a university but also mirror those same websites in Europe.16.171. in the network below the user on the Makerere campus gets http://www. network throughput of up to 22 Mbps can be achieved by using standard.makerere.ug A 172.21 A 172.11g wireless gear.13. The DNS server on the campus in the above diagram has a zone file for makerere. it gets the IP address at the African ISP.21 A 172. whereas a user elsewhere on the Internet gets it resolved to 195.Chapter 3: Network Design 89 Internet resolves the same names to different IP addresses. they get the IP address of the mirrored web server in Europe.makerere.16. Internet link optimization As mentioned earlier.ug domain. unlicensed 802.21 But there is another DNS server on the Internet that is actually authoritative for the makerere. When visitors from other countries access that website. and all computers on the campus are configured to use it as their DNS server.16.ac. as web hosting close to the Internet backbone has become very cheap. For example.ug and is configured as if it is authoritative for that domain.ug/ resolved to 172. In addition.makerere.21.13 ftp A 195.makerere.ug exchange.ac. Whenever clients of that ISP access the website.ac.ac.171.33. and so the traffic stays in the same country.16. international visitors do not congest the ISP's VSAT connection when visiting the university's website. it serves as the DNS caching server for the Makerere campus.171.16.16. In this way.16.13 mail A 16.21 MX mail.ug Split DNS is not dependent on using RFC 1918 addresses.ac.16.ac. One of the zones is used by internal network clients and the other by users on the Internet.makerere. An African ISP might.16.16.132. This amount of bandwidth will likely be at least an order of magnitude higher than that provided by .ac. This is achieved by having two zones on two different DNS servers for the same domain. for example.ac.ug www A 195.16.ug ftpserver. This is becoming an attractive solution. The DNS records for the campus DNS server would look like this: makerere.

The high latency in satellite networks is due to the long distance to the satellite and the constant speed of light. Generally speaking. large bandwidth delay product. But if your primary Internet connection is through a VSAT link. and transmission errors. this section might apply if the ISP's connection is via VSAT. This distance adds about 520 ms to a packet s round-trip time (RTT).00 t er 35 s Thousands of Kilometers Figure 3.90 Chapter 3: Network Design your Internet link.28: Due to the speed of light and long distances involved. The factors that most significantly impact TCP/IP performance are long RTT. operating systems that support modern TCP/IP implementations should be used in a satellite network. Most Internet connections in Africa and other parts of the developing world are via VSAT. s 35 t er . This term refers to factors that affect TCP/IP performance on any network that has relatively large bandwidth. you can significantly improve response times when accessing Internet hosts. These implementations support the RFC 1323 extensions: . and should be able to comfortably support many simultaneous Internet users. even if a university gets its connection via an ISP.00 me 0K 0K ilo ilo me . a single ping packet can take more than 520 ms to be acknowledged over a VSAT link. compared to a typical RTT between Europe and the USA of about 140 ms. TCP/IP factors over a satellite connection A VSAT is often referred to as a long fat pipe network. Therefore. but high latency. you will encounter some performance issues if you rely on default TCP/IP parameters. By optimizing your VSAT link.

Large bandwidth-delay product The amount of data in transit on a link at any point of time is the product of bandwidth and the RTT. Long round-trip time (RTT) Satellite links have an average RTT of around 520ms to the first hop. but when a large file is transferred acceptable data rates are achieved after a while. Because of the high latency of the satellite link. This drastically decreases the throughput of short-duration TCP connections. the remote host is always allowed to send a certain amount of data without acknowledgment. and for a satellite link it means that TCP stays in slow-start mode for a longer time than would otherwise be the case. Furthermore. However. regardless of the fact that the capacity of the link may be much greater. To utilize the link fully. the value of the bandwidth-delay product is important. TCP enters the congestion-control phase. and owing to the higher RTT.and long-duration TCP connections. This gives a maximum data rate of 123 KB/s. • Timestamps for calculating appropriate RTT and retransmission timeout values for the link in use. The advertised window is the receiver's current available buffer size. This is can be seen in the way that a small website might take surprisingly long to load.Chapter 3: Network Design 91 • The window scale option for supporting large TCP window sizes (larger than 64KB). remains in this phase for a longer time. An acknowledgment is usually required for all incoming data on a TCP/IP connection. the bandwidth-delay product is large. This amount of data is called the TCP window size. the window size of the connection should be equal to the bandwidth-delay product. Time spent in the slow-start stage is proportional to the RTT. or 64KB / 520 ms. which is 984 kbps. If the largest window size allowed is 64KB. thus reducing the throughput of both short. the maximum theoretical throughput achievable via satellite is (window size) / RTT. The window size is usually 64KB in modern TCP/IP implementations. Each TCP segment header contains a field called advertised window. On satellite networks. . TCP/IP allows the remote host to send a certain amount of data in advance without acknowledgment. which specifies how many additional bytes of data the receiver is prepared to accept. TCP uses the slow-start mechanism at the start of a connection to find the appropriate TCP/IP parameters for that connection. • Selective acknowledgment (SACK) to enable faster recovery from transmission errors. when packets are lost. which is important to achieve a good transfer rate on large bandwidthdelay product connections.

once this congestion-control phase is started. a technique known as TCP acknowledgment spoofing can be used (see Performance Enhancing Proxy. because a 64 KB window size can fill up to 984 kbps. Therefore errors on a satellite link have a more serious effect on the performance of TCP than over low latency links. mechanisms such as Selective Acknowledgment (SACK) have been developed. TCP/IP on satellite links will take a longer time to return to the previous throughput level. for example. the administrator might want those downloads to make use of the full bandwidth. This may also become critical . But if the university has more than 984 kbps. Transmission errors In older TCP/IP implementations. To overcome this limitation. there are large scheduled downloads at night. below). This buffer size has a maximum value of 64KB in most modern TCP/IP implementations. and the "long fat pipe network" factors might be an obstacle. The Microsoft Windows 2000 TCP/IP Implementation Details White Paper states "Windows 2000 introduces support for an important performance feature known as Selective Acknowledgment (SACK). To maximize performance. SACK is especially important for connections using large TCP window sizes. packet loss is always considered to have been caused by congestion (as opposed to link errors). because many people are using the bandwidth. TCP performs congestion avoidance. it might in some cases not get the full bandwidth of the available link due to the "long fat pipe network" factors discussed above. What these factors really imply is that they prevent a single machine from filling the entire bandwidth. the sender should set its send buffer size and the receiver should set its receive buffer size to no less than the bandwidth-delay product. This is not a bad thing during the day. requiring three duplicate ACKs or slow start in the case of a timeout. allowing the sender to retransmit only those segments that are missing because of link errors. When this happens." SACK has been a standard feature in Linux and BSD kernels for quite some time. Because of the long RTT value. To overcome the problem of TCP/IP stacks from operating systems that don't increase the window size beyond 64KB. But if. Implications for universities If a site has a 512 kbps connection to the Internet. the default TCP/IP settings are likely sufficient. Be sure that your Internet router and your ISP s remote side both support SACK.92 Chapter 3: Network Design The sender is not allowed to send more bytes than the advertised window. SACK specifies exactly those packets that have been received.

or purchased "off the shelf" from a number of vendors. set up comprehensive monitoring and analysis tools. and would be a proxy server with a large disk cache that has RFC 1323 extensions. More information While bandwidth optimization is a complex and often difficult subject. then the only machines that make connections to the Internet will be the proxy and mail servers. communicate using a different TCP session or even their own proprietary protocol.html . In this way. and thus the link characteristics that affect protocol performance (long fat pipe factors) are overcome (by TCP acknowledgment spoofing. for example). Such a system can be built from scratch using Squid. see the free book How to Accelerate Your Internet (http://bwmo. . A laptop has a TCP session with the PEP at the ISP. and implement a network architecture that enforces desired usage limits.psc.org/rfc/rfc3135). The PEP at the satellite provider gets the files from the web server. To make the best possible use of available bandwidth. and the one at the satellite provider. If a university has implemented a network where all traffic has to go through the proxy (enforced by network layout). Additionally. the techniques in this chapter should help reduce obvious sources of wasted bandwidth. among other features. the TCP session is split. the PEP makes use of proxying and pre-fetching to accelerate web access further.ietf. see http://www. For more information about bandwidth optimization. Administrators might consider taking steps to ensure that the full bandwidth can be achieved by tuning their TCP/IP settings. for example.Chapter 3: Network Design 93 if a significant amount of your network traffic routes through a single tunnel or VPN connection to the other end of the VSAT link.edu/networking/perf_tune. you will need to define a good access policy.net/). For more information. Performance-enhancing proxy (PEP) The idea of a Performance-enhancing proxy is described in RFC 3135 (see http://www. That PEP.

.

4
Antennas & Transmission Lines
The transmitter that generates the RF1 power to drive the antenna is usually located at some distance from the antenna terminals. The connecting link between the two is the RF transmission line. Its purpose is to carry RF power from one place to another, and to do this as efficiently as possible. From the receiver side, the antenna is responsible for picking up any radio signals in the air and passing them to the receiver with the minimum amount of distortion, so that the radio has its best chance to decode the signal. For these reasons, the RF cable has a very important role in radio systems: it must maintain the integrity of the signals in both directions. There are two main categories of transmission lines: cables and waveguides. Both types work well for efficiently carrying RF power at 2.4 GHz.

Cables
RF cables are, for frequencies higher than HF, almost exclusively coaxial cables (or coax for short, derived from the words “of common axis”). Coax cables have a core conductor wire surrounded by a non-conductive material called dielectric, or simply insulation. The dielectric is then surrounded by an encompassing shielding which is often made of braided wires. The dielectric prevents an electrical connection between the core and the shielding. Finally, the coax is protected by an outer casing which is generally made

1. Radio Frequency. See chapter two for discussion of electromagnetic waves.

95

96

Chapter 4: Antennas & Transmission Lines

from a PVC material. The inner conductor carries the RF signal, and the outer shield prevents the RF signal from radiating to the atmosphere, and also prevents outside signals from interfering with the signal carried by the core. Another interesting fact is that high frequency electrical signal always travels along the outer layer of a conductor: the larger the central conductor, the better signal will flow. This is called the “skin effect”.
Outer Jacket Shield Dielectric Conductor Figure 4.1: Coaxial cable with jacket, shield, dielectric, and core conductor.

Even though the coaxial construction is good at containing the signal on the core wire, there is some resistance to the electrical flow: as the signal travels down the core, it will fade away. This fading is known as attenuation, and for transmission lines it is measured in decibels per meter (dB/m). The rate of attenuation is a function of the signal frequency and the physical construction of the cable itself. As the signal frequency increases, so does its attenuation. Obviously, we need to minimize the cable attenuation as much as possible by keeping the cable very short and using high quality cables. Here are some points to consider when choosing a cable for use with microwave devices: 1. “The shorter the better!” The first rule when you install a piece of cable is to try to keep it as short as possible. The power loss is not linear, so doubling the cable length means that you are going to lose much more than twice the power. In the same way, reducing the cable length by half gives you more than twice the power at the antenna. The best solution is to place the transmitter as close as possible to the antenna, even when this means placing it on a tower. 2. “The cheaper the worse!” The second golden rule is that any money you invest in buying a good quality cable is a bargain. Cheap cables are intended to be used at low frequencies, such as VHF. Microwaves require the highest quality cables available. All other options are nothing but a dummy load 2.

2. A dummy load is a device that dissipates RF energy without radiating it. Think of it as a heat sink that works at radio frequencies.

Chapter 4: Antennas & Transmission Lines

97

3. Always avoid RG-58. It is intended for thin Ethernet networking, CB or VHF radio, not for microwave. 4. Always avoid RG-213. It is intended for CB and HF radio. In this case the cable diameter does not imply a high quality, or low attenuation. 5. Whenever possible, use Heliax (also called “Foam”) cables for connecting the transmitter to the antenna. When Heliax is unavailable, use the best rated LMR cable you can find. Heliax cables have a solid or tubular center conductor with a corrugated solid outer conductor to enable them to flex. Heliax can be built in two ways, using either air or foam as a dielectric. Air dielectric Heliax is the most expensive and guarantees the minimum loss, but it is very difficult to handle. Foam dielectric Heliax is slightly more lossy, but is less expensive and easier to install. A special procedure is required when soldering connectors in order to keep the foam dielectric dry and uncorrupted. LMR is a brand of coax cable available in various diameters that works well at microwave frequencies. LMR-400 and LMR-600 are a commonly used alternative to Heliax. 6. Whenever possible, use cables that are pre-crimped and tested in a proper lab. Installing connectors to cable is a tricky business, and is difficult to do properly even with the proper tools. Unless you have access to equipment that can verify a cable you make yourself (such as a spectrum analyzer and signal generator, or time domain reflectometer), troubleshooting a network that uses homemade cable can be difficult. 7. Don t abuse your transmission line. Never step over a cable, bend it too much, or try to unplug a connector by pulling directly the cable. All of those behaviors may change the mechanical characteristic of the cable and therefore its impedance, short the inner conductor to the shield, or even break the line. Those problems are difficult to track and recognize and can lead to unpredictable behavior on the radio link.

Waveguides
Above 2 GHz, the wavelength is short enough to allow practical, efficient energy transfer by different means. A waveguide is a conducting tube through which energy is transmitted in the form of electromagnetic waves. The tube acts as a boundary that confines the waves in the enclosed space. The Faraday cage effect prevents electromagnetic effects from being evident outside the guide. The electromagnetic fields are propagated through the waveguide by means of reflections against its inner walls, which are considered perfect conductors. The intensity of the fields is greatest at the center along the X dimension, and must diminish to zero at the end walls because the existence of any field parallel to the walls at the surface would cause an infinite current to flow in a perfect conductor. Waveguides, of course, cannot carry RF in this fashion.

98

Chapter 4: Antennas & Transmission Lines

The X, Y and Z dimensions of a rectangular waveguide can be seen in the following figure:
Y

Z X Figure 4.2: The X, Y, and Z dimensions of a rectangular waveguide.

There are an infinite number of ways in which the electric and magnetic fields can arrange themselves in a waveguide for frequencies above the low cutoff frequency. Each of these field configurations is called a mode. The modes may be separated into two general groups. One group, designated TM (Transverse Magnetic), has the magnetic field entirely transverse to the direction of propagation, but has a component of the electric field in the direction of propagation. The other type, designated TE (Transverse Electric) has the electric field entirely transverse, but has a component of magnetic field in the direction of propagation. The mode of propagation is identified by the group letters followed by two subscript numerals. For example, TE 10, TM 11, etc. The number of possible modes increases with the frequency for a given size of guide, and there is only one possible mode, called the dominant mode, for the lowest frequency that can be transmitted. In a rectangular guide, the critical dimension is X. This dimension must be more than 0.5 at the lowest frequency to be transmitted. In practice, the Y dimension usually is made about equal to 0.5 X to avoid the possibility of operation in other than the dominant mode. Cross-sectional shapes other than the rectangle can be used, the most important being the circular pipe. Much the same considerations apply as in the rectangular case. Wavelength dimensions for rectangular and circular guides are given in the following table, where X is the width of a rectangular guide and r is the radius of a circular guide. All figures apply to the dominant mode.

Chapter 4: Antennas & Transmission Lines

99

Type of guide
Cutoff wavelength Longest wavelength transmitted with little attenuation Shortest wavelength before next mode becomes possible

Rectangular
2X

Circular
3.41r

1.6X

3.2r

1.1X

2.8r

Energy may be introduced into or extracted from a waveguide by means of either an electric or magnetic field. The energy transfer typically happens through a coaxial line. Two possible methods for coupling to a coaxial line are using the inner conductor of the coaxial line, or through a loop. A probe which is simply a short extension of the inner conductor of the coaxial line can be oriented so that it is parallel to the electric lines of force. A loop can be arranged so that it encloses some of the magnetic lines of force. The point at which maximum coupling is obtained depends upon the mode of propagation in the guide or cavity. Coupling is maximum when the coupling device is in the most intense field. If a waveguide is left open at one end, it will radiate energy (that is, it can be used as an antenna rather than as a transmission line). This radiation can be enhanced by flaring the waveguide to form a pyramidal horn antenna. We will see an example of a practical waveguide antenna for WiFi later in this chapter. Cable Type
RG-58 RG-213 LMR-400 3/8” LDF

Core
0.9 mm 2.26 mm 2.74 mm 3.1 mm

Dielectric
2.95 mm 7.24 mm 7.24 mm 8.12 mm

Shield
3.8 mm 8.64 mm 8.13 mm 9.7 mm

Jacket
4.95 mm 10.29 mm 10.29 mm 11 mm

Here is a table contrasting the sizes of various common transmission lines. Choose the best cable you can afford with the lowest possible attenuation at the frequency you intend to use for your wireless link.

100

Chapter 4: Antennas & Transmission Lines

Connectors and adapters
Connectors allow a cable to be connected to another cable or to a component of the RF chain. There are a wide variety of fittings and connectors designed to go with various sizes and types of coaxial lines. We will describe some of the most popular ones. BNC connectors were developed in the late 40s. BNC stands for Bayonet Neill Concelman, named after the men who invented it: Paul Neill and Carl Concelman. The BNC product line is a miniature quick connect / disconnect connector. It features two bayonet lugs on the female connector, and mating is achieved with only a quarter turn of the coupling nut. BNC's are ideally suited for cable termination for miniature to subminiature coaxial cable (RG58 to RG-179, RG-316, etc.) They have acceptable performance up to few GHz. They are most commonly found on test equipment and 10base2 coaxial Ethernet cables. TNC connectors were also invented by Neill and Concelman, and are a threaded variation of the BNC. Due to the better interconnect provided by the threaded connector, TNC connectors work well through about 12 GHz. TNC stands for Threaded Neill Concelman. Type N (again for Neill, although sometimes attributed to “Navy”) connectors were originally developed during the Second World War. They are usable up to 18 Ghz, and very commonly used for microwave applications. They are available for almost all types of cable. Both the plug / cable and plug / socket joints are waterproof, providing an effective cable clamp. SMA is an acronym for SubMiniature version A, and was developed in the 60s. SMA connectors are precision, subminiature units that provide excellent electrical performance up to 18 GHz. These high-performance connectors are compact in size and mechanically have outstanding durability. The SMB name derives from SubMiniature B, and it is the second subminiature design. The SMB is a smaller version of the SMA with snap-on coupling. It provides broadband capability through 4 GHz with a snap-on connector design. MCX connectors were introduced in the 80s. While the MCX uses identical inner contact and insulator dimensions as the SMB, the outer diameter of the plug is 30% smaller than the SMB. This series provides designers with options where weight and physical space are limited. MCX provides broadband capability though 6 GHz with a snap-on connector design.

Chapter 4: Antennas & Transmission Lines

101

In addition to these standard connectors, most WiFi devices use a variety of proprietary connectors. Often, these are simply standard microwave connectors with the center conductor parts reversed, or the thread cut in the opposite direction. These parts are often integrated into a microwave system using a short jumper called a pigtail that converts the non-standard connector into something more robust and commonly available. Some of these connectors include: RP-TNC. This is a TNC connector with the genders reversed. These are most commonly found on Linksys equipment, such as the WRT54G. U.FL (also known as MHF). The U.FL is a patented connector made by Hirose, while the MHF is a mechanically equivalent connector. This is possibly the smallest microwave connector currently in wide use. The U.FL / MHF is typically used to connect a mini-PCI radio card to an antenna or larger connector (such as an N or TNC). The MMCX series, which is also called a MicroMate, is one of the smallest RF connector line and was developed in the 90s. MMCX is a micro-miniature connector series with a lock-snap mechanism allowing for 360 degrees rotation enabling flexibility. MMCX connectors are commonly found on PCMCIA radio cards, such as those manufactured by Senao and Cisco. MC-Card connectors are even smaller and more fragile than MMCX. They have a split outer connector that breaks easily after just a few interconnects. These are commonly found on Lucent / Orinoco / Avaya equipment. Adapters, which are also called coaxial adapters, are short, two-sided connectors which are used to join two cables or components which cannot be connected directly. Adapters can be used to interconnect devices or cables with different types. For example, an adapter can be used to connect an SMA connector to a BNC. Adapters may also be used to fit together connectors of the same type, but which cannot be directly joined because of their gender.

Figure 4.3: An N female barrel adapter.

For example a very useful adapter is the one which enables to join two Type N connectors, having socket (female) connectors on both sides.

102

Chapter 4: Antennas & Transmission Lines

Choosing the proper connector
1. “The gender question.” Virtually all connectors have a well defined gender consisting of either a pin (the “male” end) or a socket (the “female” end). Usually cables have male connectors on both ends, while RF devices (i.e. transmitters and antennas) have female connectors. Devices such as directional couplers and line-through measuring devices may have both male and female connectors. Be sure that every male connector in your system mates with a female connector. 2. “Less is best!” Try to minimize the number of connectors and adapters in the RF chain. Each connector introduces some additional loss (up to a few dB for each connection, depending on the connector!) 3. “Buy, don t build!” As mentioned earlier, buy cables that are already terminated with the connectors you need whenever possible. Soldering connectors is not an easy task, and to do this job properly is almost impossible for small connectors as U.FL and MMCX. Even terminating “Foam” cables is not an easy task. 4. Don t use BNC for 2.4 GHz or higher. Use N type connectors (or SMA, SMB, TNC, etc.) 5. Microwave connectors are precision-made parts, and can be easily damaged by mistreatment. As a general rule, you should rotate the outer sleeve to tighten the connector, leaving the rest of the connector (and cable) stationary. If other parts of the connector are twisted while tightening or loosening, damage can easily occur. 6. Never step over connectors, or drop connectors on the floor when disconnecting cables (this happens more often than what you may imagine, especially when working on a mast over a roof). 7. Never use tools like pliers to tighten connectors. Always use your hands. When working outside, remember that metals expand at high temperatures and reduce their size at low temperatures: a very tightened connector in the summer can bind or even break in winter.

Antennas & radiation patterns
Antennas are a very important component of communication systems. By definition, an antenna is a device used to transform an RF signal traveling on a conductor into an electromagnetic wave in free space. Antennas demonstrate a property known as reciprocity, which means that an antenna will maintain the same characteristics regardless if whether it is transmitting or receiving. Most antennas are resonant devices, which operate efficiently over a relatively narrow frequency band. An antenna must be tuned to the same frequency band of the radio system to which it is connected, otherwise

If the antenna has an impedance different than 50 . then there is a mismatch and an impedance matching circuit is required. the impedance of the radio. a high return loss will yield unacceptable antenna performance.Chapter 4: Antennas & Transmission Lines 103 the reception and the transmission will be impaired. Bandwidth The bandwidth of an antenna refers to the range of frequencies over which the antenna can operate correctly.FL FC Bandwidth = 100 . antenna. there are a few common terms that must be defined and explained: Input Impedance For an efficient transfer of energy. When any of these components are mismatched. The bandwidth can also be described in terms of percentage of the center frequency of the band. The antenna's bandwidth is the number of Hz for which the antenna will exhibit an SWR less than 2:1. Antenna term glossary Before we talk about specific antennas. Return loss Return loss is another way of expressing mismatch. It is a logarithmic ratio measured in dB that compares the power reflected by the antenna to the power that is fed into the antenna from the transmission line. and transmission cable connecting them must be the same. transmission efficiency suffers. The relationship between SWR and return loss is the following: Return Loss (in dB) = 20log10 SWR SWR-1 While some energy will always be reflected back into the system. the antenna will emit radiation distributed in space in a certain way. Transceivers and their transmission lines are typically designed for 50 impedance. F H . When a signal is fed into an antenna. A graphical representation of the relative distribution of the radiated power in space is called a radiation pattern.

If a wireless link uses fixed locations for both ends.. bandwidth is constant relative to frequency. it is possible to use antenna directivity to concentrate the radiation beam in the wanted direction. but it is a dimensionless ratio.where FH is the highest frequency in the band. is technically known as a gain transfer technique.104 Chapter 4: Antennas & Transmission Lines .. Any real antenna will radiate more energy in some directions than in others. The two most common reference antennas are the isotropic antenna and the resonant half-wave dipole antenna. which has a calibrated gain. Directivity and Gain Directivity is the ability of an antenna to focus energy in a particular direction when transmitting. To compare the dipole to an antenna over a range of frequencies requires a number of dipoles of different lengths. In a mobile application where the transceiver is not fixed. Real isotropic antennas do not exist. the total power radiated is the same as an isotropic antenna. The isotropic antenna radiates equally well in all directions. which is the gain in the direction in which the antenna is radiating most of the power. or to receive energy from a particular direction when receiving. The resonant half-wave dipole can be a useful standard for comparing to other antennas at one frequency or over a very narrow band of frequencies. Different types of antennas have different bandwidth limitations. and FC is the center frequency in the band. An antenna gain of 3 dB compared to a dipole antenna would be written as 3 dBd. and so the antenna should ideally radiate as well as possible in all directions. Since antennas cannot create energy. but they provide useful and simple theoretical antenna patterns with which to compare real antennas. Usually we are only interested in the maximum gain. The method of measuring gain by comparing the antenna under test against a known standard antenna. An antenna gain of 3 dB compared to an isotropic antenna would be written as 3 dBi. FL is the lowest frequency in the band. Another method for measuring gain is the 3 anten- . Gain is not a quantity which can be defined in terms of a physical quantity such as the Watt or the Ohm. In this way. The gain of an antenna in a given direction is the amount of energy radiated in that direction compared to the energy an isotropic antenna would radiate in the same direction when driven with the same input power. Gain is given in reference to a standard antenna. it may be impossible to predict where the transceiver will be. An omnidirectional antenna is used in these applications. Any additional energy radiated in the directions it favors is offset by equally less energy radiated in all other directions. it would be different depending upon the center frequency. If bandwidth was expressed in absolute units of frequency.

The radiation pattern is threedimensional. in the horizontal or vertical planes. dB -5 -10 -15 -20 -25 -30 -35 -40 -45 -50 -180° -140° -100° -60° -20° 20° 60° 100° 140° 180° Figure 4. The following figure shows a rectangular plot presentation of a typical ten-element Yagi. The detail is good but it is difficult to visualize the antenna behavior in different directions. the equally spaced concentric circles may be replaced with appropriately placed circles representing the decibel response. can also be plotted on a linear coordinate system. For ease of comparison. In this kind of plot the minor lobes are suppressed. Such a grid may be used to prepare a linear plot of the power contained in the signal. These pattern measurements are presented in either a rectangular or a polar format. points are located by projection along a rotating axis (radius) to an intersection with one of several concentric circles. where the transmitted and received power at the antenna terminals is measured between three arbitrary antennas at a known fixed distance. This grid enhances plots in which the antenna has a high directivity and small minor lobes. . too. referenced to 0 dB at the outer edge of the plot. but usually the measured radiation patterns are a twodimensional slice of the three-dimensional pattern. at a constant distance. In the linear coordinate system. In the polar-coordinate graph. rather than the power.4: A rectangular plot of a yagi radiation pattern. In this case. Radiation Pattern The radiation pattern or antenna pattern describes the relative strength of the radiated field in various directions from the antenna. and are graduated. Polar coordinate systems are used almost universally. since it also describes the receiving properties of the antenna. The voltage of the signal.Chapter 4: Antennas & Transmission Lines 105 nas method. The following is a polar plot of the same 10 element Yagi antenna. The radiation pattern is a reception pattern as well. the concentric circles are equally spaced. Lobes with peaks more than 15 dB or so below the main lobe disappear because of their small size. Polar coordinate systems may be divided generally in two classes: linear and logarithmic.

In the logarithmic polar coordinate system the concentric grid lines are spaced periodically according to the logarithm of the voltage in the signal.106 Chapter 4: Antennas & Transmission Lines the directivity is enhanced and the minor lobes suppressed. and the gain transfer method is then used to establish the absolute gain of the antenna. but not in the same degree as in the linear power grid. This is shown in Figure 4. There are two kinds of radiation pattern: absolute and relative. which is greater than the spacing between -50 dB and -53 dB. Generally the 0 dB reference for the outer edge of the chart is used.5: A linear polar plot of the same yagi. lobes that are 30 or 40 dB below the main lobe are still distinguishable. and this choice will have an effect on the appearance of the plotted patterns. Different values may be used for the logarithmic constant of periodicity. The spacing thus correspond to the relative significance of such changes in antenna performance. Absolute radiation patterns are presented in absolute units of field strength or power. Relative radiation patterns are referenced in relative units of field strength or power. The spacing between points at 0 dB and at -3 dB is greater than the spacing between -20 dB and -23 dB. Most radiation pattern measurements are relative to the isotropic antenna. 0° 270° 90° 180° Figure 4. A modified logarithmic scale emphasizes the shape of the major beam while compressing very low-level (>30 dB) sidelobes towards the center of the pattern. With this type of grid.6. .

and so antenna patterns are usually measured in the far-field region. For pattern measurement it is important to choose a distance sufficiently large to be in the far-field.Chapter 4: Antennas & Transmission Lines 0° 107 270° 90° 180° Figure 4. The angular distance between the half power points is defined as the beamwidth. The term near-field refers to the field pattern that exists close to the antenna. well out of the near-field. The far-field is also called the radiation field. The minimum permissible distance depends on the dimensions of the antenna in relation to the wavelength. and then the points on either side of the peak which represent half the power of the peak intensity are located. Ordinarily. The accepted formula for this distance is: rmin = 2d2 where rmin is the minimum distance from the antenna. and is the wavelength. Half the power expressed in decibels is -3dB. . and is what is most commonly of interest. d is the largest dimension of the antenna.6: The logarithmic polar plot The radiation pattern in the region close to the antenna is not the same as the pattern at large distances. while the term far-field refers to the field pattern at large distances. it is the radiated power that is of interest. Beamwidth An antenna's beamwidth is usually understood to mean the half-power beamwidth. so the half power beamwidth is sometimes referred to as the 3dB beamwidth. The peak radiation intensity is found. Both horizontal and vertical beamwidth are usually considered.

a horizontal orientation. the null is useful for several purposes. the directive gain increases. both of which are perpendicular to the direction of propagation. such as suppression of interfering signals in a given direction. Nulls In an antenna radiation pattern. With linear polarization. a null is a zone in which the effective radiated power is at a minimum. Thus. The initial polarization of a radio wave is determined by the antenna. the electric field vector stays in the same plane all the time. Omnidirectional antennas always have vertical polarization. Polarization Polarization is defined as the orientation of the electric field of an electromagnetic wave. Sidelobes No antenna is able to radiate all the energy in one preferred direction. direction of propagation electric field magnetic field Figure 4. Some is inevitably radiated in other directions. A null often has a narrow directivity angle compared to that of the main beam. such reflections cause variations in received signal strength. Two special cases of elliptical polarization are linear polarization and circular polarization. commonly specified in dB down from the main lobe. The electric field may leave the antenna in a vertical orientation.7: The electrical wave is perpendicular to magnetic wave. These smaller peaks are referred to as sidelobes. With horizontal polarization. Horizontal antennas are less likely to pick up man-made interference. then the directive gain is inversely proportional to the beamwidth: as the beamwidth decreases. . or at some angle between the two.108 Chapter 4: Antennas & Transmission Lines Assuming that most of the radiated power is not divided into sidelobes. Vertically polarized radiation is somewhat less affected by reflections over the transmission path. which ordinarily is vertically polarized. Polarization is in general described by an ellipse.

. This is the ratio of the maximum directivity of an antenna to its directivity in the opposite direction. which can be determined using the following formula: Loss (dB) = 20 log (cos ) . for 30° we lose 1. In the real world. Front-to-back ratio It is often useful to compare the front-to-back ratio of directional antennas. This technique can sometimes be used to build stable links..3dB. For 15° the loss is approximately 0. the greater the mismatch in polarization between a transmitting and receiving antenna. such as yagis or can antennas. a 90° mismatch in polarization is quite large but not infinite. when the radiation pattern is plotted on a relative dB scale. You can use the polarization effect to your advantage on a point-to-point link.. This reduction in power transfer will reduce the overall system efficiency and performance. there will be a reduction in power transfer between the two antennas. Use a monitoring tool to observe interference from adjacent networks.where is the difference in alignment angle between the two antennas.25dB. This rotation may be right-hand or left-hand. Polarization Mismatch In order to transfer maximum power between a transmit and a receive antenna.Chapter 4: Antennas & Transmission Lines 109 In circular polarization the electric field vector appears to be rotating with circular motion about the direction of propagation. physical antenna misalignment will result in a polarization mismatch loss. the greater the apparent loss. When the transmit and receive antennas are both linearly polarized. making one full turn for each RF cycle. the same polarization sense. Choice of polarization is one of the design choices available to the RF system designer. and rotate one antenna until you see the lowest received signal. When the antennas are not aligned or do not have the same polarization. For example. Some antennas. In short. the front-to-back ratio is the difference in dB between the level of the maximum radiation in the forward direction and the level of radiation at 180 degrees. and the same axial ratio. Then bring your link online and orient the other end to match polarization. for 45° we lose 3dB and for 90° we have an infinite loss. even in noisy radio environments. both antennas must have the same spatial orientation. can be simply rotated 90° to match the polarization of the other end of the link.

The most popular types of omnidirectional antennas are the dipole and the ground plane. Types of directive antennas are the Yagi. At 2. or as narrow as 60 degrees. so the antennas must be different in size to radiate signals at the correct wavelength. or sectorial antennas which focus into a small area.5 cm. another classification can be used: • Application. • Physical construction. Directional or directive antennas are antennas in which the beamwidth is much narrower than in sectorial antennas. to coffee cans. • Directivity. the parabolic dish. while remote links are point-to-point. A brief list of common type of antennas for the 2. the biquad. The wavelength is different at different frequencies. They have the highest gain and are therefore used for long distance links.110 Chapter 4: Antennas & Transmission Lines This number is meaningless for an omnidirectional antenna. while at 5 GHz it is 6 cm. Types of Antennas A classification of antennas can be based on: • Frequency and size. Each of these suggest different types of antennas for their purpose.4 GHz the wavelength is 12. with a short description and basic information about their characteristics. Nodes that are used for multipoint access will likely use omni antennas which radiate equally in all directions. Sectorial antennas radiate primarily in a specific area. We are particularly interested in antennas working in the microwave range. especially in the 2. the horn. the helicoidal. The beam can be as wide as 180 degrees. to parabolic dishes. Omnidirectional antennas radiate roughly the same pattern all around the antenna in a complete 360° pattern. Antennas can be omnidirectional. ranging from simple wires. Directive antennas are the primary choice for this application. Antennas used for HF are different from antennas used for VHF. In the point-to-point case. the patch antenna. antennas are used to connect two single locations together.4 GHz WLAN use. and many others. Antennas can be constructed in many different ways.4 GHz and 5 GHz frequencies. sectorial or directive. When considering antennas suitable for 2. Access points tend to make point-to-multipoint networks. but it gives you an idea of the amount of power directed forward on a very directional antenna. which in turn are different from antennas for microwave. .4 GHz frequency is presented now.

The more directors a Yagi has. the greater the gain. The antenna propagates electromagnetic field energy in the direction running from the driven element toward the directors. and approximately 0. The vertical beamwidth represents the degree of flatness in the focus.Chapter 4: Antennas & Transmission Lines 111 1/4 wavelength ground plane The 1⁄4 wavelength ground plane antenna is very simple in its construction and is useful for communications when size. This set of elements. each measuring approximately half wavelength. and is most sensitive to incoming electromagnetic field energy in this same direction. half-wave dipole antenna. The gain of this antenna is in the order of 2 . This is useful in a Point-to-Multipoint situation. called radials. Yagi antenna A basic Yagi consists of a certain number of straight elements. Figure 4. To increase the gain. if all the other antennas are also at the same height.2 to 0. and providing more focus on the horizon. As more direc- . The driven or active element of a Yagi is the equivalent of a center-fed. a director is placed in front of the driven element and is slightly shorter than half wavelength. is known as a ground plane.8: Quarter wavelength ground plane antenna. It consists of a 1⁄4 wave element as half-dipole and three or four 1⁄4 wavelength ground elements bent 30 to 45 degrees down. A reflector is placed behind the driven element and is slightly longer than half wavelength. Parallel to the driven element. or simply passive elements.4 dBi. This is a simple and effective antenna that can capture a signal equally from all directions. the signal can be flattened out to take away focus from directly above and below. cost and ease of construction are important. are straight rods or wires called reflectors and directors.5 wavelength on either side of it. A typical Yagi has one reflector and one or more directors. This antenna is designed to transmit a vertically polarized signal.

15 dBi.112 Chapter 4: Antennas & Transmission Lines tors are added to a Yagi. At 2. Figure 4. a simple horn antenna made with a tin can has a gain in the order of 10 . rather than a dipole antenna or any other type of antenna. Horn antennas are commonly used as the active element in a dish antenna. The direction of maximum radiation corresponds with the axis of the horn. but can be fed with a coaxial cable and a proper transition. Yagi antennas are used primarily for Point-to-Point links. The horn is pointed toward the center of the dish reflector. The flared portion can be square.4 GHz.10: Feed horn made from a food can. Horn The horn antenna derives its name from the characteristic flared appearance. The use of a horn. have a gain from 10 to 20 dBi and a horizontal beamwidth of 10 to 20 degrees. cylindrical or conical. Figure 4. it therefore becomes longer. Following is the photo of a Yagi antenna with 6 directors and one reflector.9: A Yagi antenna. rectangular. It is easily fed with a waveguide. at the focal point of the dish minimizes loss of energy around the edges of the dish reflector. .

BiQuad The BiQuad antenna is simple to build and offers good directivity and gain for Point-to-Point communications. if the squares are placed side by side the polarization is vertical.11: A solid dish antenna. . aluminum. Aluminum is frequently used for its weight advantage. but are safer to use and easier to build. Dishes up to one meter are usually made from solid material. The main advantage is that they can be made to have gain and directivity as large as required. The polarization is such that looking at the antenna from the front. brass. Copper. It can be used as stand-alone antenna or as feeder for a Parabolic Dish. its durability and good electrical characteristics.Chapter 4: Antennas & Transmission Lines 113 Parabolic Dish Antennas based on parabolic reflectors are the most common type of directive antennas when a high gain is required. This antenna has a beamwidth of about 70 degrees and a gain in the order of 10-12 dBi. It consists of a two squares of the same size of 1⁄4 wavelength as a radiating element and of a metallic plate or grid as reflector. Figure 4. Dishes which have a reflecting surface that uses an open mesh are frequently used. These have a poorer front-to-back ratio. Windage increases rapidly with dish size and soon becomes a severe problem. The main disadvantage is that big dishes are difficult to mount and are likely to have a large windage. galvanized steel and iron are suitable mesh materials.

The position of the focus. all the energy received by the dish from a distant source is reflected to a single point at the focus of the dish. Other Antennas Many other types of antennas exist and new ones are created following the advances in technology. with a gain up to 20 dB. Their horizontal beamwidth can be as wide as 180 degrees. • Sector or Sectorial antennas: they are widely used in cellular telephony infrastructure and are usually built adding a reflective plate to one or more phased dipoles. • Panel or Patch antennas: they are solid flat panels used for indoor coverage. or focal length. while the vertical is usually much narrower. Reflector theory The basic property of a perfect parabolic reflector is that it converts a spherical wave irradiating from a point source placed at the focus into a plane wave. is given by: f = 16 D2 c . Composite antennas can be built with many Sectors to cover a wider horizontal range (multisectorial antenna).114 Chapter 4: Antennas & Transmission Lines Figure 4.12: The BiQuad.. or as narrow as 60 degrees. Conversely.where D is the dish diameter and c is the depth of the parabola at its center.. .

. The gain and beamwidth obtained are given by: Gain = ( 2 D)2 n Beamwidth = 70 D . simply adding amplifiers will not magically solve all of your networking problems.Chapter 4: Antennas & Transmission Lines 115 The size of the dish is the most important factor since it determines the maximum gain that can be achieved at the given frequency and the resulting beamwidth. Each time the diameter of a dish is doubled. but also by other factors. and quickly power up and amplify the signal. Amplifiers are available that work at 2. These devices sense when an attached radio is transmitting.where D is the dish diameter and n is the efficiency. An efficiency of 50% can be assumed when hand-building the antenna. and can add several Watts of power to your transmission. or 6 dB.25 corresponds to the common focal-plane dish in which the focus is in the same plane as the rim of the dish. they also add amplification to the signal before sending it to the radio. antennas do not actually create power. Amplifiers must work at relatively wide bandwidths at 2. and has an additional lead that connects to a power source.. Amplifiers As mentioned earlier. Unfortunately.4 GHz. If both stations double the size of their dishes. By using a power amplifier.4 GHz. They then switch off again when transmission ends. a very substantial gain. The efficiency is determined mainly by the effectiveness of illumination of the dish by the feed. you can use DC power to augment your available signal. The ratio is directly related to the beamwidth of the feed necessary to illuminate the dish effectively. the gain is four times. They simply direct all available power into a particular pattern. signal strength can be increased of 12 dB. When receiving. An amplifier connects between the radio transmitter and the antenna. The ratio f / D (focal length/diameter of the dish) is the fundamental factor governing the design of the feed for a dish. greater. We do not discuss power amplifiers at length in this book because there are a number of significant drawbacks to using them: • They are expensive. and must switch quickly enough to work for Wi-Fi applications.. The value of 0. Two dishes of the same diameter but different focal lengths require different design of feed if both are to be illuminated efficiently.

• They provide no additional directionality. so must everyone else on the block. By using higher gain antennas rather than amplifiers. Adding an antenna to a highly amplified signal will likely cause the link to exceed legal limits. but what happens when the neighbors decide to do the same thing with their radios? Using amplifiers for a wireless link causes roughly the same effect at 2. Your link may “work better” for the moment. Adding antenna gain provides both gain and directionality benefits to both ends of the link. This approach may scale to exactly one user. and can make interference problems worse. but they tend to cost several hundred dollars per unit. and can improve a link simply by changing the antenna on one end.4 GHz antennas has fallen dramatically since the introduction of 802. This may not be much of an issue today in rural areas. you avoid all of these problems. you are creating a louder source of noise for other users of the unlicensed band. adding antenna gain will improve your link and can actually decrease the noise level for your neighbors. but tend to reject noise from other directions.116 Chapter 4: Antennas & Transmission Lines These amplifiers do exist. • Amplifiers generate noise for other users of the band. and so turns it up to full volume. it will likely be able to be heard but will not be able to hear the other end. but you will have problems when other users of the band decide to use amplifiers of their own. Practical antenna designs The cost of 2. and so we recommend pursuing them before adding amplifiers.11b. Amplifiers blindly amplify both desired and interfering signals.4 GHz. • Using amplifiers probably isn t legal. amplifiers work best at amplifying a transmitted signal. These techniques are unlikely to cause problems for other users of the band. Conversely. They not only improve the available amount of signal. Using amplifiers is often compared to the inconsiderate neighbor who wants to listen to the radio outside their home. Innovative designs use simpler parts and fewer materials to achieve . • You will need at least two. Every country imposes power limits on use of unlicensed spectrum. They might even “improve” reception by pointing their speakers out the window. By increasing your output power. Antennas cost far less than amps. If you only add an amplifier to one end of a link with insufficient antenna gain. Whereas antennas provide reciprocal gain that benefits both sides of a connection. While they may now be able to hear the radio. but it can cause big problems in populated areas. Using more sensitive radios and good quality cable also helps significantly on long distance shots.

By placing the internal dipole antenna present in USB wireless dongles at the focus of a parabolic dish. requiring just a piece of wire. To test the antenna. You will need to waterproof the dongle and cable if the antenna is used outdoors. The plate has a hole drilled in the mid- . You should see a significant improvement in gain (20 dB or more) when you find the proper position.net. Many USB-fed parabolic designs and ideas are documented online at http://www. an N socket and a square metallic plate. and even metal cookware (such as a wok. television antennas. Many kinds of parabolic dishes will work. constructing antennas from locally available components is very straightforward. availability of good antennas is still limited in many areas of the world. you can provide significant gain without the need to solder or even open the wireless device itself. or strainer).orcon. Using the laptop s client driver or a tool such as Netstumbler (see Chapter 6). or simply try the dongle in both positions to see which provides more gain. you will need to find the orientation and location of the dipole inside the dongle. Try various positions while watching your signal strength meter until you find the optimum location. Collinear omni This antenna is very simple to build. inexpensive and lossless USB cable is then used to feed the antenna. You can either open the dongle and look for yourself. observe the received signal strength of the access point. Now.usbwifi. Unfortunately. To build a USB dongle parabolic. As a bonus. eliminating the need for expensive coaxial cable or Heliax. It can be used for indoor or outdoor pointto-multipoint short distance coverage.nz/ . USB dongle as dish feed Possibly the simplest antenna design is the use of a parabola to direct the output of a USB wireless device (known in networking circles as a USB dongle). slowly move the dongle in relation to the parabolic while watching the signal strength meter. point it at an access point several meters away. Once the best location is found. round lid. Most devices orient the dipole to be parallel with the short edge of the dongle. including satellite dishes. and can be a lot of fun. Use a silicone compound or a piece of PVC tubing to seal the electronics against the weather. The proper position will vary depending on the shape of the parabola and the construction of the USB dongle. and importing them can be prohibitively expensive.Chapter 4: Antennas & Transmission Lines 117 impressive gain with relatively little machining. securely fix the dongle in place. but some will mount the dipole perpendicular to the short edge. While actually designing an antenna can be a complex and error-prone process. and connect the USB dongle to a laptop. We present four practical antenna designs that can be built for very little money.

118 Chapter 4: Antennas & Transmission Lines dle to accommodate an N type chassis socket that is screwed into place. The wire is soldered to the center pin of the N socket and has coils to separate the active phased elements. Parts list and tools required • One screw-on N-type female connector • 50 cm of copper or brass wire of 2 mm of diameter • 10x10 cm or greater square metallic plate Figure 4. • Ruler • Pliers • File • Soldering iron and solder • Drill with a set of bits for metal (including a 1. while the long one with four elements will have 7 to 9 dBi of gain.5 cm diameter bit) • A piece of pipe or a drill bit with a diameter of 1 cm • Vice or clamp • Hammer • Spanner or monkey wrench Construction 1. Straighten the wire using the vice.13: 10 cm x 10 cm aluminum plate. Two versions of the antenna are possible: one with two phased elements and two coils and another with four phased elements and four coils. . We are going to describe how to build the long antenna only. For the short antenna the gain will be around 5 dBi.

bend once again the wire over this second line at 90 degrees.14: Make the wire as straight as you can.15: Gently tap the wire to make a sharp bend. draw a line at 2.6 cm from the bend. On this line. With a marker. in the opposite direction to the first bend but in the same plane. The wire should look like a Z . 3. .5 cm starting from one end of the wire. 2.Chapter 4: Antennas & Transmission Lines 119 Figure 4. Using the vice and the hammer. bend the wire at 90 degrees with the help of the vice and of the hammer. Draw another line at a distance of 3. Figure 4.

4. The coil will look like this: Figure 4.8 cm from the first one. we will use the pipe or the drill bit and curve the wire around it. 5. We will now twist the Z portion of the wire to make a coil with a diameter of 1 cm. with the help of the vice and of the pliers. You should make a second coil at a distance of 7.18: The completed coil.120 Chapter 4: Antennas & Transmission Lines Figure 4. To do this. Make a third and a fourth coil following the . Figure 4.16: Bend the wire into a “Z” shape.17: Bend the wire around the drill bit to make a coil. Both coils should have the same turning direction and should be placed on the same side of the wire.

8 cm one from each other.19: Try to keep it as straight possible. 6. If the coils have been made correctly. With a marker and a ruler. Figure 4. finding its center. . Increase the diameter of the hole using bits with an increasing diameter. make a pilot hole at the center of the plate. Trim the last phased element at a distance of 8. Figure 4.0 cm from the fourth coil. With a small diameter drill bit. at the same distance of 7. it should now be possible to insert a pipe through all the coils as shown.Chapter 4: Antennas & Transmission Lines 121 same procedure.20: Inserting a pipe can help to straighten the wire. draw the diagonals on the metallic plate.

The hole should fit the N connector exactly. Figure 4.5 cm of copper pipe with an external diameter of 2 cm.122 Chapter 4: Antennas & Transmission Lines Figure 4. and place it between the connector and the plate. cut 0.23: Adding a copper pipe spacer helps to match the impedance of the antenna to 50 Ohms. Figure 4.21: Drilling the hole in the metal plate. it is important that the visible surface of the internal insulator of the connector (the white area around the central pin) is at the same level as the surface of the plate. Use a file if needed.22: The N connector should fit snugly in the hole. To have an antenna impedance of 50 Ohms. 7. For this reason. .

Keeping the wire vertical with the pliers. Smooth with the file the side of the wire which is 2. The first coil should be at 3. Tin the wire for around 0.Chapter 4: Antennas & Transmission Lines 123 8.0 cm from the plate. With the soldering iron. . from the first coil. 9. Screw the nut to the connector to fix it firmly on the plate using the spanner.24: Secure the N connector tightly to the plate. Figure 4. Figure 4. 10. solder its tinned side in the hole of the central pin.25: Add a little solder to the end of the wire to “tin” it prior to soldering. tin the central pin of the connector.5 cm long.5 cm at the smoothed end helping yourself with the vice.

26: The first coil should start 3. 11. Be very gentle and try not to scrape the surface of the wire with the pliers. Figure 4.0 cm. . Using the use the vice and the pliers. Repeat the same procedure for the other three coils.27: Stretching the coils.0 cm. stretching their length to 2. We are now going to stretch the coils extending the total vertical length of the wire. 12.124 Chapter 4: Antennas & Transmission Lines Figure 4. you should pull the cable so that the final length of the coil is of 2.0 cm from the surface of the plate.

you can check the curve of the reflected power of the antenna.5 cm from the plate to the top. At the end the antenna should measure 42.Chapter 4: Antennas & Transmission Lines 125 Figure 4. 13.28: Repeat the stretching procedure for all of the remaining coils.5 cm from the plate to the end of the wire. The picture below shows the display of the spectrum analyzer.29: The finished antenna should be 42. . 14. Figure 4. If you have a spectrum analyzer with a tracking generator and a directional coupler.

1. where G .126 Chapter 4: Antennas & Transmission Lines Figure 4. so the can diameter should be in the range of 7. and seal the antenna shut with silicone or PVC glue.60 and 0. sometimes called a Cantenna.3 .75 wavelength in air at the design frequency. Cut a hole at the bottom for the transmission line. The acceptable values for the diameter D of the feed are between 0. juice. or other tin can.44 GHz the wavelength is 12. uses a tin can as a waveguide and a short wire soldered on an N connector as a probe for coaxial-cable-to-waveguide transition. useful for short to medium distance point-to-point links.9. It may be also used as a feeder for a parabolic dish or grid. It is a directional antenna. If you intend to use this antenna outside. 2.2 cm. Cantenna The waveguide antenna. The simplest method is to enclose the whole thing in a large piece of PVC pipe closed with caps.75 is the guide wavelength and is given by: G . recycling a food.30: A spectrum plot of the reflected power of the collinear omni. you will need to weatherproof it. The length L of the can preferably should be at least 0.2 cm. At 2. It can be easily built at just the price of the connector. Not all cans are good for building an antenna because there are dimensional constraints.

( / 1. which at 2.31: Dimensional constraints on the cantenna The gain for this antenna will be in the order of 10 to 14 dBi. The probe for coaxial cable to waveguide transition should be positioned at a distance S from the bottom of the can. with a beamwidth of around 60 degrees.8 cm. L S D Figure 4.2 cm we need a can of at least 14. 3.706D)2) For D = 7.3 cm and a height of about 21 cm. For our example. the longer the can should be.4 cm.Chapter 4: Antennas & Transmission Lines 127 G = sqrt(1 . . Generally the smaller the diameter.25 G Its length should be 0.25 . while for D = 9. given by: S = 0.05 cm. we need a can of at least 56.44 GHz corresponds to 3.3 cm. we will use oil cans that have a diameter of 8.

128 Chapter 4: Antennas & Transmission Lines Figure 4. Parts list • one screw-on N-type female connector • 4 cm of copper or brass wire of 2 mm of diameter • an oil can of 8.32: The finished cantenna. .33: Parts needed for the can antenna.3 cm of diameter and 21 cm of height Figure 4.

2.34: Be careful of sharp edges when opening the can. The circular disk has a very sharp edge.Chapter 4: Antennas & Transmission Lines 129 Tools required • Can opener • Ruler • Pliers • File • Soldering iron • Solder • Drill with a set of bits for metal (with a 1. Be careful to measure from the inner side of the bottom. Figure 4. With the can opener. or some other tasty treat. Use a punch (or a small drill bit or a Phillips screwdriver) and a hammer to mark the point. carefully remove the upper part of the can. measure 6. have a friend serve the food. Be careful not to . cookies.2 cm from the bottom of the can and draw a point.5 cm diameter bit) • Vice or clamp • Spanner or monkey wrench • Hammer • Punch Construction 1. With the ruler. This makes it easier to precisely drill the hole. If the can contained pineapple. Be careful when handling it! Empty the can and wash it with soap.

Figure 4.5 cm at the same end helping yourself with the vice. Tin the wire for around 0. Figure 4. make a hole at the center of the plate. 4. 3. With a small diameter drill bit. Increase the diameter of the hole using bits with an increasing diameter. Use the file to smooth the border of the hole and to remove the painting around it in order to ensure a better electrical contact with the connector. then use a larger bit to finish the job. The hole should fit exactly the N connector.35: Mark the hole before drilling. . Smooth with the file one end of the wire.130 Chapter 4: Antennas & Transmission Lines change the shape of the can doing this by inserting a small block of wood or other object in the can before tapping it.36: Carefully drill a pilot hole.

37: Tin the end of the wire before soldering. Trim the wire at 3. Insert a washer and gently screw the nut onto the connector.05 cm measured from the bottom part of the nut. 5. . 6. solder its tinned side in the hole of the central pin. Figure 4. With the soldering iron.Chapter 4: Antennas & Transmission Lines 131 Figure 4.38: Solder the wire to the gold cup on the N connector. tin the central pin of the connector. Keeping the wire vertical with the pliers.

You are done! . Figure 4. 7.39: The length of the wire is critical. Unscrew the nut from the connector.132 Chapter 4: Antennas & Transmission Lines Figure 4. Use the pliers or the monkey wrench to screw firmly the nut on the connector. Insert the connector into the hole of the can.40: Assemble the antenna. 8. Screw the nut on the connector from inside the can. leaving the washer in place.

without using an amplifier. Use the technique described in the USB dongle antenna example (watching signal strength changes over time) to find the optimum location of the can for the dish you are using. and seal the ends with caps and glue. PVC works well for the can antenna. With very large parabolas. Details about this achievement can be found at http://www. By using a well-built cantenna with a properly tuned parabolic. in 2005. Cantenna as dish feed As with the USB dongle parabolic. The link crossed a distance of over 200 kilometers! The wireless enthusiasts used a 3. a team of college students successfully established a link from Nevada to Utah in the USA. As with the other antenna designs.wifi-shootout. you can achieve an overall antenna gain of 30dBi or more. you should make a weatherproof enclosure for the antenna if you wish to use it outdoors. You will need to drill a hole in the side of the tube to accommodate the N connector on the side of the can.5 meter satellite dish to establish an 802.Chapter 4: Antennas & Transmission Lines 133 Figure 4.11b link that ran at 11 Mbps. so does the potential gain and directivity of the antenna. Insert the entire can in a large PVC tube. you can use the cantenna design as a feeder for significantly higher gain. As the size of the parabolic increases.com/ . For example. you can achieve significantly higher gain. Mount the can on the parabolic with the opening of the can pointed at the center of the dish.41: Your finished cantenna.

NEC2 lets you build an antenna model in 3D. It is also worth installing the xnecview package for structure verification and radiation pattern plotting.nittany-scientific. The next important element is the character of the ground. but it is an invaluable tool for antenna designers. and how the wires are connected up. The conductivity of the earth varies from place to place. type nec2 and enter the input and output filenames. Once these are defined. It is a complex tool and requires some research to learn how to use it effectively. The controls tell NEC where the RF source is connected. This can be broken up into various sections. but in many cases it plays a vital role in determining the antenna gain pattern. A frequency or range of frequencies of the RF signal must be specified. the transmitting antenna is then modeled. It was developed more than ten years ago and has been compiled to run on many different computer systems. but also has some surface patch modeling capability. An antenna described in NEC2 is given in two parts: its structure and a sequence of controls. NEC2 is available from http://www. To run NEC2 on Linux. and how we can modify the design in order to get the maximum gain.134 Chapter 4: Antennas & Transmission Lines NEC2 NEC2 stands for Numerical Electromagnetics Code (version 2) and is a free antenna modeling package. Windows and Mac versions are also available.nec2. install the NEC2 package from the URL below. You should see the expected pattern. The advantage of NEC2 is that we can get an idea of how the antenna works before building it. NEC2 is particularly effective for analyzing wiregrid models. but for a quick idea of what it represents a gain pattern can be plotted using xnecview. and then the model is built using this text description. with a peak at the optimum angle of takeoff.org/ Online documentation can be obtained from the "Unofficial NEC Home Page" at http://www. . If all went well you should have a file containing the output. horizontally omnidirectional. and then analyzes the antenna s electromagnetic response. Because of the reciprocity theorem the transmitting gain pattern is the same as the receiving one. The antenna design is described in a text file.com/nec/ . The structure is simply a numerical description of where the different parts of the antenna are located. so modeling the transmission characteristics is sufficient to understand the antenna's behavior completely. To launch it.

Note that this diagram is not to scale. we ll look at the sort of features and attributes that are desirable in a wireless component. nor is it necessarily the best choice of network design. In this chapter. an unprecedented surge in interest in wireless networking hardware has brought a huge variety of inexpensive equipment to the market. You obviously need at least one computer connected to an Ethernet network. So much variety. but along the way they may need to interface with an amplifier. Many components require power.5 Networking Hardware In the last couple of years. that it would be impossible to catalog every available component. All of these components use various sorts of connectors. Radio components need to be connected to antennas. and you may well be wondering why this stuff is referred to as “wireless”. in fact. and a wireless router or bridge attached to the same network. Wired wireless With a name like “wireless”. 135 . and see several examples of commercial and DIY gear that has worked well in the past. you may be surprised at how many wires are involved in making a simple point-to-point link. A wireless node consists of many components. But it will introduce you to many common interconnects and components that you will likely encounter in the real world. either via an AC mains line or using a DC transformer. Now multiply those cables and connectors by the number of nodes you will bring online. not to mention a wide variety of cable types and thicknesses. The diagram on the next page will give you some idea of the cabling required for a typical point-to-point link. lightning arrestor. or other device. which must all be connected to each other with appropriate cabling.

DC Transformer Wireless to Ethernet bridge LMR-400 feed line N Connectors Chapter 5: Networking Hardware CAT5 carrying data and DC power to AC Power POE Injector UPS Amplifier 24 dBi Parabolic dish 8 dBi Omni RP-TNC Connectors Access Point Pigtail adapter cable DC Transformer Lightning arrestor UPS Hello Joe! [ 10+ Kilometers ] Ethernet switch Hello Joe! to AC Power 136 Ethernet switch Figure 5. .1: Component interconnects.

• Range. establishing the scope of your project is critical before any purchasing decisions are made. is this an important factor for this segment of your network? If the gear in question supports an open protocol (such as 802. 4. be sure to consider these variables: • Interoperability. Will the equipment you are considering work with equipment from other manufacturers? If not. but should not be considered an absolute truth. An antenna that is connected via feed line. and other factors. determining the available budget. 2. bridge. range is not something inherent in a particular piece of equipment. Electrical components consisting of power supplies. Choosing wireless components Unfortunately. The actual selection of hardware should be determined by establishing the requirements for the project. the surrounding terrain. The old saying that “you get what you pay for” often holds true when buying high tech equipment. An existing computer or network connected to an Ethernet switch. it is more useful to know the transmission power of the radio as well as the antenna gain (if . and lightning arrestors. Rather than relying on a semifictional “range” rating supplied by the manufacturer.Chapter 5: Networking Hardware 137 While the actual components used will vary from node to node. A device that connects that network to a wireless device (a wireless router. the characteristics of the device at the other end of the link. in a world of competitive hardware manufacturers and limited budgets. then it will likely interoperate with equipment from other sources. conditioners. it is vital to understand precisely what you get for your money so you can make a choice that fits your needs. and verifying that the project is feasible using the available resources (including providing for spares and ongoing maintenance costs). As discussed in Chapter 1. every installation will incorporate these parts: 1. or is integrated into the wireless device itself. 3. When comparing wireless equipment for use in your network. the price tag is the single factor that usually receives the most attention. A device s range depends on the antenna connected to it. As we saw in Chapter 4.11b/g). or repeater). While the price tag is an important part of any purchasing decision.

does the device include an external antenna connector? If so. a good rule of thumb is to divide the device “speed” by two. Keep in mind that the radio symbol rate (eg. This can be used as a measure of the quality of the hardware. It is unlikely that you will be able to answer every possible question before buying gear. cables. power over Ethernet injectors typically are not. • Required accessories. To keep the initial price tag low. When in doubt. or other features critical to the intended network design? By answering these questions first. perform throughput testing on an evaluation unit before committing to purchasing a large amount of equipment that has no official throughput rating. should your project require it? What is the projected life span of this particular product. What about pigtails. a lower number is better for radio sensitivity. at least at the fastest and slowest speeds. what is the cost to increase these limits? What is the physical form factor of the device? How much power does it consume? Does it support POE as a power source? Does the device provide encryption. vendors often leave out accessories that are required for normal use. what type is it? Are there user or throughput limits imposed by software. but if you prioritize the questions and press the vendor to answer them before committing to a purchase. you will make the best use of your budget and build a network of components that are well suited to your needs. you will be able to make intelligent buying decisions when it comes time to choose networking hardware. With this information. you can calculate the theoretical range as described in Chapter 3. . Does the price tag include all power adapters? (DC supplies are typically included. 54 Mbps) is never the actual throughput rating of the device (eg. Manufacturers consistently list the highest possible bit rate as the “speed” of their equipment. Double-check input voltages as well. Will you be able to easily replace failed components? Can you order the part in large quantity. as well as allow you to complete a link budget calculation. as equipment is often provided with a US-centric power supply). • Radio sensitivity. As we saw in Chapter 3. both in terms of useful running time in-the-field and likely availability from the vendor? • Other factors. • Throughput. and subtract 20% or so. does the device include a weatherproof case? • Availability. and radio cards? If you intend to use it outdoors. For example. bandwidth monitoring tools.11g). Be sure that other needed features are provided for to meet your particular needs. NAT. How sensitive is the radio device at a given bit rate? The manufacturer should supply this information. about 22 Mbps for 802. adapters. antennas.138 Chapter 5: Networking Hardware an antenna is included). and if so. If throughput rate information is not available for the device you are evaluating.

and many offer support and warranty for an extended period for a monthly fee. robust solutions for very low cost. but they can also allow relatively unskilled people to install and maintain equipment. such as antennas and towers. The art of building local telecommunications infrastructure lies in finding the right balance of money to effort needed to be expended to solve the problem at hand. If a piece of equipment simply doesn t work or is difficult to configure or troubleshoot. The chief strengths of commercial solutions are that they provide support and a (usually limited) equipment warranty. By using off-the-shelf radio cards. by importing technology.Chapter 5: Networking Hardware 139 Commercial vs. In many regions. Of course. At this stage of human technology. motherboards. This is not to say that commercial equipment is inferior to a do-it-yourself solution. there is a limit to how much work can be done by any individual or group in a given amount of time. DIY solutions Your network project will almost certainly consist of components purchased from vendors as well as parts that are sourced or even fabricated locally. importing every component needed to build a network is prohibitively expensive for all but the largest budgets. such as a lightning strike) then the manufacturer will typically replace it. Most will provide these services for a limited time as part of the purchase price. are likely far too complex to consider having them fabricated locally. and only importing components that must be purchased. Combining open hardware platforms with open source software can yield significant “bang for the buck” by providing custom. are relatively simple and can be made locally for a fraction of the cost of importing. By providing so-called “turn-key solutions”. and other components. you can exchange money for equipment that can solve a particular problem in a comparatively short amount of time. Between these extremes lie the communication devices themselves. To put it another way. This is a basic economic truth in most areas of the world. By providing a consistent . global distribution of information is quite trivial compared to global distribution of goods. You can save considerable money in the short term by finding local sources for parts and labor. Should the equipment fail in normal use (barring extreme damage. Some components. you can build devices that provide features comparable (or even superior) to most commercial implementations. They also provide a consistent platform that tends to lead to very stable. such as radio cards and antenna feed line. Other components. a good manufacturer will assist you. manufacturers not only save development time. often interchangeable network installations.

140

Chapter 5: Networking Hardware

platform, it is simple to keep spares on hand and simply “swap out” equipment that fails in the field, without the need for a technician to configure equipment on-site. Of course, all of this comes at comparatively higher initial cost for the equipment compared to off-the-shelf components. From a network architect s point of view, the three greatest hidden risks when choosing commercial solutions are vendor lock-in, discontinued product lines, and ongoing licensing costs. It can be costly to allow the lure of ill-defined new “features” drive the development of your network. Manufacturers will frequently provide features that are incompatible with their competition by design, and then issue marketing materials to convince you that you simply cannot live without them (regardless of whether the feature contributes to the solution of your communications problem). As you begin to rely on these features, you will likely decide to continue purchasing equipment from the same manufacturer in the future. This is the essence of vendor lock-in. If a large institution uses a significant amount of proprietary equipment, it is unlikely that they will simply abandon it to use a different vendor. Sales teams know this (and indeed, some rely on it) and use vendor lock-in as a strategy for price negotiations. When combined with vendor lock-in, a manufacturer may eventually decide to discontinue a product line, regardless of its popularity. This ensures that customers, already reliant on the manufacturer s proprietary features, will purchase the newest (and nearly always more expensive) model. The long term effects of vendor lock-in and discontinued products are difficult to estimate when planning a networking project, but should be kept in mind. Finally, if a particular piece of equipment uses proprietary computer code, you may need to license use of that code on an ongoing basis. The cost of these licenses may vary depending on features provided, number of users, connection speed, or other factors. If the license fee is unpaid, some equipment is designed to simply stop working until a valid, paid-up license is provided! Be sure that you understand the terms of use for any equipment you purchase, including ongoing licensing fees. By using generic equipment that supports open standards and open source software, you can avoid some of these pitfalls. For example, it is very difficult to become locked-in to a vendor that uses open protocols (such as TCP/IP over 802.11a/b/g). If you encounter a problem with the equipment or the vendor, you can always purchase equipment from a different vendor that will interoperate with what you have already purchased. It is for these reasons that we recommend using proprietary protocols and licensed spectrum only in cases where the open equivalent (such as 802.11a/b/g) is not technically feasible.

Chapter 5: Networking Hardware

141

Likewise, while individual products can always be discontinued at any time, you can limit the impact this will have on your network by using generic components. For example, a particular motherboard may become unavailable on the market, but you may have a number of PC motherboards on hand that will perform effectively the same task. We will see some examples of how to use these generic components to build a complete wireless node later in this chapter. Obviously, there should be no ongoing licensing costs involved with open source software (with the exception of a vendor providing extended support or some other service, without charging for the use of the software itself). There have occasionally been vendors who capitalize on the gift that open source programmers have given to the world by offering the code for sale on an ongoing licensed basis, thereby violating the terms of distribution set forth by the original authors. It would be wise to avoid such vendors, and to be suspicious of claims of “free software” that come with an ongoing license fee. The disadvantage of using open source software and generic hardware is clearly the question of support. As problems with the network arise, you will need to solve those problems for yourself. This is often accomplished by consulting free online resources and search engines, and applying code patches directly. If you do not have team members who are competent and dedicated to designing a solution to your communications problem, then it can take a considerable amount of time to get a network project off the ground. Of course, there is never a guarantee that simply “throwing money at the problem” will solve it either. While we provide many examples of how to do much of the work yourself, you may find this work very challenging. You will need to find the balance of commercial solutions and the do-ityourself approach that works for project. In short, always define the scope of your network first, identify the resources you can bring to bear on the problem, and allow the selection of equipment to naturally emerge from the results. Consider commercial solutions as well as open components, while keeping in mind the long-term costs of both. When considering which equipment to use, always remember to compare the expected useful distance, reliability, and throughput, in addition to the price. Be sure to include any ongoing license fees when calculating the overall cost of the equipment. And finally, make sure that the radios you purchase operate in an unlicensed band where you are installing them, or if you must use licensed spectrum, that you have budget and permission to pay for the appropriate licenses.

142

Chapter 5: Networking Hardware

Professional lightning protection
Lightning is a natural predator of wireless equipment. There are two different ways lightning can strike or damage equipment: direct hits or induction hits. Direct hits happen when lightning actually hits the tower or antenna. Induction hits are caused when lightning strikes near the tower. Imagine a negatively charged lightning bolt. Since like charges repel each other, that bolt will cause the electrons in the cables to move away from the strike, creating current on the lines. This can be much more current than the sensitive radio equipment can handle. Either type of strike will usually destroy unprotected equipment.

Figure 5.2: A tower with a heavy copper grounding wire.

Protecting wireless networks from lightning is not an exact science, and there is no guarantee that a lightning strike will not happen, even if every single precaution is taken. Many of the methods used will help prevent both direct and induction strikes. While it is not necessary to use every single lightning protection method, using more methods will help further protect the equipment. The amount of lightning historically observed within a service area will be the biggest guide to how much needs to be done. Start at the very bottom of the tower. Remember, the bottom of the tower is below the ground. After the tower foundation is laid, but before the hole is backfilled, a ring of heavy braided ground wire should have been installed with the lead extending above ground surfacing near a tower leg. The wire should be American Wire Gauge (AWG) #4 or thicker. In addition, a backup

Chapter 5: Networking Hardware

143

ground or earthing rod should be driven into the ground, and a ground wire run from the rod to the lead from the buried ring. It is important to note that not all steel conducts electricity the same way. Some types of steel act as better electrical conductors then others, and different surface coatings can also affect how tower steel handles electrical current. Stainless steel is one of the worst conductors, and rust proof coatings like galvanizing or paint lessen the conductivity of the steel. For this reason, a braided ground wire is run from the bottom of the tower all the way to the top. The bottom needs to be properly attached to the leads from both the ring and the backup ground rod. The top of the tower should have a lightning rod attached, and the top of that needs to be pointed. The finer and sharper the point, the more effective the rod will be. The braided ground wire from the bottom needs to be terminated at this grounding rod. It is very important to be sure that the ground wire is connected to the actual metal. Any sort of coating, such as paint, must be removed before the wire is attached. Once the connection is made, the exposed area can be repainted, covering the wire and connectors if necessary to save the tower from rust and other corrosion. The above solution details the installation of the basic grounding system. It provides protection for the tower itself from direct hits, and installs the base system to which everything else will connect. The ideal protection for indirect induction lightning strikes are gas tube arrestors at both ends of the cable. These arrestors need to be grounded directly to the ground wire installed on the tower if it is at the high end. The bottom end needs to be grounded to something electrically safe, like a ground plate or a copper pipe that is consistently full of water. It is important to make sure that the outdoor lightning arrestor is weatherproofed. Many arresters for coax cables are weatherproofed, while many arresters for CAT5 cable are not. In the event that gas arrestors are not being used, and the cabling is coax based, then attaching one end of a wire to the shield of the cable and the other to the ground wire installed on the towers will provide some protection. This can provide a path for induction currents, and if the charge is weak enough, it will not affect the conductor wire of the cable. While this method is by no means as good of protection as using the gas arrestors, it is better then doing nothing at all.

Building an access point from a PC
Unlike consumer operating systems (such as Microsoft Windows), the GNU/ Linux operating system gives a network administrator the potential for full access to the networking stack. One can access and manipulate network

144

Chapter 5: Networking Hardware

packets at any level from the data-link layer through the application layer. Routing decisions can be made based on any information contained in a network packet, from the routing addresses and ports to the contents of the data segment. A Linux-based access point can act as a router, bridge, firewall, VPN concentrator, application server, network monitor, or virtually any other networking role you can think of. It is freely available software, and requires no licensing fees. GNU/Linux is a very powerful tool that can fill a broad variety of roles in a network infrastructure. Adding a wireless card and Ethernet device to a PC running Linux will give you a very flexible tool that can help you deliver bandwidth and manage your network for very little cost. The hardware could be anything from a recycled laptop or desktop machine to an embedded computer, such as a Linksys WRT54G or Metrix networking kit. In this section we will see how to configure Linux in the following configurations: • As a wireless access point with Masquerading/NAT and a wired connection to the Internet (also referred to as a wireless gateway). • As a wireless access point that acts as a transparent bridge. The bridge can be used either as a simple access point, or as a repeater with 2 radios. Consider these recipes as a starting point. By building on these simple examples, you can create a server that fits precisely into your network infrastructure.

Prerequisites
Before proceeding, you should already be familiar with Linux from a users perspective, and be capable of installing the Gnu/Linux distribution of your choice. A basic understanding of the command line interface (terminal) in Linux is also required. You will need a computer with one or more wireless cards already installed, as well as a standard Ethernet interface. These examples use a specific card and driver, but there are a number of different cards that should work equally well. Wireless cards based on the Atheros and Prism chipsets work particularly well. These examples are based on Ubuntu Linux version 5.10 (Breezy Badger), with a wireless card that is supported by the HostAP or MADWiFi drivers. For more information about these drivers, see http://hostap.epitest.fi/ and http://madwifi.org/ . The following software is required to complete these installations. It should be provided in your Linux distribution:

Chapter 5: Networking Hardware

145

• Wireless Tools (iwconfig, iwlist commands) • iptables firewall • dnsmasq (caching DNS server and DHCP server) The CPU power required depends on how much work needs to be done beyond simple routing and NAT. For many applications, a 133MHz 486 is perfectly capable of routing packets at wireless speeds. If you intend to use a lot of encryption (such as WEP or a VPN server), then you will need something faster. If you also want to run a caching server (such as Squid) then you will need a computer with plenty of fast disk space and RAM. A typical router that is only performing NAT will operate will with as little as 64MB of RAM and storage. When building a machine that is intended to be part of your network infrastructure, keep in mind that hard drives have a limited lifespan compared to most other components. You can often use solid state storage, such as a flash disk, in place of a hard drive. This could be a USB flash drive (assuming your PC will boot from USB), or a Compact Flash card using a CF to IDE adapter. These adapters are quite inexpensive, and will make a CF card appear act like standard IDE hard drive. They can be used in any PC that supports IDE hard drives. Since they have no moving parts, they will operate for many years through a much wider range of temperatures than a hard disk will tolerate.

Scenario 1: Masquerading access point
This is the simplest of the scenarios, and is especially useful in situations where you want a single access point for an office setting. This is easiest in a situation where: 1. There is an existing dedicated firewall and gateway running Linux, and you just want to add a wireless interface. 2. You have an old refurbished computer or laptop available, and prefer to use that as an access point. 3. You require more power in terms of monitoring, logging and/or security than most commercial access points provide, but don't want to splurge on an enterprise access point. 4. You would like a single machine to act as 2 access points (and firewall) so that you can offer both a secure network access to the intranet, as well as open access to guests.

146

Chapter 5: Networking Hardware

Initial setup
Start of with an already configured computer running GNU/Linux. This could be an Ubuntu Server installation, or Fedora Core. The computer must have at least 2 interfaces for this to work, and at least one of these interfaces should be wireless. The rest of this description assumes that your cabled Ethernet port (eth0) is connected to the Internet, and that there is a wireless interface (wlan0) that will provide the access point functionality. To find out if your chipset supports master mode, try the following command as root:
# iwconfig wlan0 mode Master

...replacing wlan0 with the name of your interface. If you get an error message, then your wireless card doesn t support access point mode. You can still try the same setup in Ad-hoc mode, which is supported by all chipsets. This requires that you to set all the laptops that are connecting to this “access point” into Ad-hoc mode as well, and may not work quite the way you are expecting. It is usually better to find a wireless card that will support AP mode. See the HostAP and MADWiFi websites mentioned earlier for a list of supported cards. Before continuing, make sure dnsmasq is installed on your machine. You can use the graphical package manager of your distribution to install it. In Ubuntu you can simply run the following as root:
# apt-get install dnsmasq

Setting up the interfaces
Set up your server so that eth0 is connected to the Internet. Use the graphical configuration tool that came with your distribution. If your Ethernet network uses DHCP, you could try the following command as root:
# dhclient eth0

You should receive an IP address and default gateway. Next, set your wireless interface to Master mode and give it a name of your choice:
# iwconfig wlan0 essid “my network” mode Master enc off

The enc off switch turns off WEP encryption. To enable WEP, add a hexkey string of the correct length:

Chapter 5: Networking Hardware

147

# iwconfig wlan0 essid “my network” mode Master enc 1A2B3C4D5E

Alternately, you can use a readable string by starting with “s:”
# iwconfig wlan0 essid “my network” mode Master enc “s:apple”

Now give your wireless interface an IP address in a private subnet, but make sure it is not the same subnet as that of your Ethernet adapter:
# ifconfig wlan0 10.0.0.1 netmask 255.255.255.0 broadcast 10.0.0.255 up

Setting up masquerading in the kernel
In order for us to be able to translate addresses between the two interfaces on the computer, we need to enable masquerading (NAT) in the linux kernel. First we load the relevant kernel module:
# modprobe ipt_MASQUERADE

Now we will flush all existing firewall rules to ensure that the firewall is not blocking us from forwarding packets between the two interfaces. If you have an existing firewall running, make sure you know how to restore the existing rules later before proceeding.
# iptables -F

Enable the NAT functionality between the two interfaces
# iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADE

Finally we need to enable the kernel to forward packets between interfaces:
# echo 1 > /proc/sys/net/ipv4/ip_forward

On Debian-based Linux distributions such as Ubuntu, this change can also be made by editing the file /etc/network/options, and be sure that ip_forward is set to yes:
ip_forward=yes

and then restarting the network interfaces with:
# /etc/init.d/network restart

or
# /etc/init.d/networking restart

html.and make sure it reads: interface=wlan0 . it provides a caching DNS server as well as a DHCP server. Then find the line that starts with: #dhcp-range= Uncomment the line and edit it to suit the match addresses being used. we will set up a DHCP server to automatically hand out addresses to wireless clients. Having a caching DNS server is especially helpful if your Internet connection is a high-latency and/or low-bandwidth connection. We use the program dnsmasq for this purpose.148 Chapter 5: Networking Hardware Setting up the DHCP server At this point we actually should have a working access point. such as a VSAT or dial-up. Install dnsmasq with your distributions package manager.. To get the basic DHCP server up and running we just need to uncomment and/or edit two lines. If dnsmasq is not available as a package.conf. In order to make it easier for people to connect to the server without knowing the IP address range.0/24 if you followed the examples). If you have enabled WEP. be sure to use the same key that you specified on the AP. /etc/dnsmasq. It can be tested by connecting to the wireless network “my network” with a separate machine and giving that machine an address in the same address range as our wireless interface on the server (10.e. It means that many DNS queries can be resolved locally. It is available from http://www.changing wlan0 to match name of your wireless interface.. and has many options for various types of configuration.0. As the name indicates. . and also making the connection feel noticeably faster for those connecting..thekelleys. Find the lines that starts: interface= . The configuration file is well commented.. saving a lot of traffic on the Internet connection.uk/dnsmasq/doc. download the source code and install it manually.org. All that is required for us to run dnsmasq is to edit a few lines of the dnsmasq configuration file. This program was developed especially for use with firewalls performing NAT. i.0.

or for an access point connected to an Ethernet. you can add extra firewall rules using whatever firewall tool is included in your distribution.a graphical client for Gnome. There are also frontends for shorewall. One inter- . you should now be able to connect to the server as an access point.0. For example. but slightly complex graphical tool that will let you create iptables scripts on a machine separate from your server.255. This should let you connect to the Internet through the server.110.255. such as webmin-shorewall • fwbuilder . your changes will continue to work should the machine need to be rebooted. central firewall and perhaps authentication server.0. Once everything is configured properly. We use a bridge instead of routing when we want both interfaces on the access point to share the same subnet. but use two wired Ethernet interfaces instead of one wired and one wireless. make sure that all settings are reflected in the system startup scripts. Because all clients share the same subnet they.0.10.6h Then save the file and start dnsmasq: # /etc/init. This does not require you to be running a graphical desktop on the server.0. This way. and get an IP address using DHCP. Adding extra security: Setting up a Firewall Once this is set up and tested.10. which requires that your server is running KDE • Shorewall – a set of scripts and configuration files that will make it easier to setup an iptables firewall.Chapter 5: Networking Hardware 149 dhcp-range=10.d/dnsmasq start That's it. This can be particularly useful in networks with multiple access points where we prefer to have a single. which requires that your server is running Gnome • knetfilter – a graphical client for KDE.0. and is a strong option for the security conscious.255.a powerful. you could setup a server as the first scenario. Some typical front-ends for setting up firewall rules include: • firestarter . can easily be managed with a single DHCP server and firewall without the need for DHCP relay. and then transfer them to the server later. Scenario 2: Transparent Bridging access point This scenario can either be used for a two-radio repeater.

This example assumes you are building a wireless repeater with two wireless interfaces. Setting up the Interfaces On Ubuntu or Debian the network interfaces are configured by editing the file /etc/network/interfaces. Follow the initial setup instructions from the previous example.150 Chapter 5: Networking Hardware face would be your Internet connection.168. Since all clients share the same subnet. and everyone will pass through the same firewall and use the same DHCP server. the bridge-utils package is required for bridging. and wlan1 will create a network called “repeater”.0.255.168. In addition. as well as for Fedora Core. but change the names of interfaces and the IP addresses accordingly. but as the number of clients increases. This package exists for Ubuntu and other Debian-based distributions.0.0.0 up pre-up ifconfig wlan1 0.1. without the requirement of dnsmasq. The simplicity of bridging comes at a cost of efficiency. and the other would connect to a switch. The IP address and netmask must match that of your existing network.168. more wireless bandwidth will be wasted on broadcast network traffic.0 up pre-up iwconfig wlan0 essid “office” mode Managed pre-up iwconfig wlan1 essid “repeater” mode Master bridge_ports wlan0 wlan1 post-down ifconfig wlan1 down post-down ifconfig wlan0 down . Add a section like the following.0 netmask 255.255 gateway 192. wlan0 and wlan1.168.1. Make sure it is installed and that the command brctl is available before proceeding.1. Then connect as many access points as you require to the same switch.1. broadcast traffic will be repeated throughout the network. Initial setup The initial setup for a bridging access point is similar to that of a masquerading access point.2 network 192. This is usually fine for small networks.0.0 broadcast 192. The wlan0 interface will be a client to the “office” network. set them up as transparent bridges.255.1 pre-up ifconfig wlan 0 0. Add the following to /etc/network/interfaces: auto br0 iface br0 inet static address 192.

255 #route add default gw 192. On Fedora Core (i.168. non-debian distributions) you still need to give your bridge interface an ip address and add a default route to the rest of the network: #ifconfig br0 192.0 broadcast 192.255.1.0 up pre-up iwconfig wlan0 essid “office” mode Managed pre-up iwconfig wlan1 essid “repeater” mode Master pre-up brctl addbr br0 pre-up brctl addif br0 wlan0 pre-up brctl addif br0 wlan1 post-down ifconfig wlan1 down post-down ifconfig wlan0 down post-down brctl delif br0 wlan0 post-down brctl delif br0 wlan1 post-down brctl delbr br0 Starting the bridge Once the bridge is defined as an interface.e.1 You should now be able to connect a wireless laptop to this new access point. The documentation for these scripts is found in /usr/share/doc/bridge-utils/.0 up pre-up ifconfig wlan1 0.1. and the details of actually setting up the bridge are handled by a couple of scripts: /etc/network/if-pre-up.0.168.0. If those scripts don't exist on your distribution (such as Fedora Core).d/bridge and /etc/network/if-post-down.168. Use the brctl command to see what your bridge is doing: # brctl show br0 .2 netmask 255.0.Chapter 5: Networking Hardware 151 Comment out any other sections in the file that refer to wlan0 or wlan1 to make sure that they don't interfere with our setup. here is an alternative setup for /etc/network/interfaces which will achieve the same thing with only marginally more hassle: iface br0 inet static pre-up ifconfig wlan 0 0.d/bridge. and connect to the Internet (or at least to the rest of your network) through this box.1. This syntax for setting up bridges via the interfaces file is specific to Debian-based distributions.255. starting the bridge is as simple as typing: # ifup -v br0 The “-v” means verbose output and will give you information to what is going on.0.

org/). By simply flashing one of these APs with the Freifunk firmware. Using Linux gives you significantly more control over how packets are routed through your network. We will see more about captive portal technology and ticketing systems in the Authentication section in Chapter 6. When the ticket expires. Siemens SE505. but can be implemented using free software such as Chillispot and phpMyPrePaid. you may wish to use a dedicated Linux distribution that is specially tailored for this purpose. you could start with either of the above two examples and implement a private wireless network where users are authenticated using a standard web browser. For example. the Freifunk firmware brings easy OLSR support to MIPS-based consumer access points. and allows for features that simply aren t possible on consumer grade access point hardware. It is maintained by Sven Ola of the Freifunk wireless group in Berlin. a Windows domain server accessible via RADIUS). the user must purchase another. In this model. you can rapidly build a self-forming OLSR mesh. and others. such as the Linksys WRT54G / WRT54GS / WAP54G. This ticketing feature is only available on relatively expensive commercial networking equipment. These are intended to be used on repurposed PCs or other networking hardware (rather than on a laptop or server) and are fine-tuned for building wireless networks. As you can see. “Wireless-friendly operating systems” for more information. wireless users can be checked against credentials in an existing database (say. This ticket provides a password that is valid for a limited amount of time (typically one day). users must purchase a ticket before accessing the network. Using a captive portal such as Chillispot. .net/wiki/FreifunkFirmware . This arrangement could allow for preferential access to users in the database. Some of these projects include: • Freifunk. See the following section. Wireless-friendly operating systems There are a number of open source operating system that provide useful tools for working with wireless networks. You can download the firmware from http://www. Freifunk is not currently available for x86 architecture machines.152 Chapter 5: Networking Hardware Scenario 1 & 2 the easy way Instead of setting up your computer as an access point from scratch. it is straightforward to provide access point services from a standard Linux router. These distributions can make the job as simple as booting from a particular CD on a computer with a wireless interface.freifunk. Another popular application is the prepaid commercial model. while providing a very limited level of access for the general public. Based on the OpenWRT project (http://openwrt.

and as such does not include userspace tools (it is not even possible to log into the machine over the network). Tomato (http://www. Some popular alternative firmware packages for the WRT54G are DD-Wrt (http://www. and OLSR.net/).freifunk. All of these distributions are designed to fit in machines with limited storage. m0n0wall is a very tiny but complete firewall package that provides AP services. and an 802. Some of these new features include client radio mode support. Pyramid is distributed and maintained by Metrix Communication LLC. network hackers realized that the firmware that shipped with the WRT54G was actually a version of Linux. Back in 2003. Unfortunately. • m0n0wall. particularly those with a background in FreeBSD.net/. It is configured through a simple web interface. If you are using a very large flash disk or hard drive. without installing unnecessary packages. it is a popular choice for wireless networkers. a four port Ethernet switch. As of this writing. and has a simple web interface for configuring networking interfaces. Based on FreeBSD. While it is not designed as an outdoor solution.com/tomato) and Freifunk (http://www. The Linksys WRT54G One of the most popular consumer access points currently on the market is the Linksys WRT54G. captive portals. you will save yourself considerable time and effort.Chapter 5: Networking Hardware 153 • Pyramid Linux.dd-wrt.polarcloud. Its goal is to provide a secure firewall. It will likely take a fair amount of development time to be sure all needed tools are included. the WRT54G sells for about $60. It is configured from a web interface and the entire system configuration is stored in a single XML file. you can certainly install a more complete OS (such as Ubuntu or Debian) and use the machine as a router or access point. in the fall of 2005.com/). making it very difficult to run Linux (it ships with VxWorks. and mesh networking.m0n0. it can be installed in a large sprinkler box or plastic tub for relatively little cost. This access point features two external RP-TNC antenna connectors. and is available at http://pyramid. This hardware revision eliminated some RAM and flash storage on the motherboard. port forwarding. This led to a tremendous interest in building custom firmware that extended the capabilities of the router significantly. It supports several different wireless cards. WifiDog. Its tiny size (less than 6MB) makes it attractive for use in very small embedded systems.ch/ . OpenWRT (http://openwrt. Pyramid is a Linux distribution for use on embedded platforms that evolved out of the venerable Pebble Linux platform. Despite this limitation. a much . Linksys released version 5 of the WRT54G.metrix. By using one of these projects as a starting point for building a wireless node. You can download m0n0wall from http://www.org/).11b/g radio.

com/wiki/index. various captive portals. and it is unclear how long it will be offered for sale. Linksys also released the WRT54GL. see http://linksysinfo. Several versions of the firmware are available from the DD-WRT website. The default login for a fresh DD-WRT installation is root with the password admin.dd-wrt. Fortunately.org/portal/content/view/27/36/ For more information about the current state of Linksys wireless router hacking. and much more. the hardware specifications may change at any time. While the WRT54GL is guaranteed to run Linux.org/ DD-WRT One popular alternate firmware for the Linksys family of access point hardware is DD-WRT (http://www. wireless hackers have now been able to install custom firmware on the notoriously difficult WRT54G version 5 and 6. It is difficult to know which hardware revision is used without opening the packaging. making it risky to purchase them at a retail store and practically impossible to order online.154 Chapter 5: Networking Hardware smaller operating system that does not allow easy customization). While these also have relatively low price tags. and the latest revisions as well(v7 and v8).scorpiontek. QoS support. which is essentially the WRT54G v4 (which runs Linux) with a slightly bigger price tag. For details on getting alternate firmware installed on a v5 or v6 access point see: http://www. and also provides SSH and telnet access. The general procedure for upgrading is to download the version of the firmware appropriate for your hardware. see the installation guide on the DD-WRT wiki at http://www. A number of other Linksys access points also run Linux. including the WRT54GS and WAP54G. the La Fonera. It uses an intuitive webbased configuration tool (unencrypted or via HTTPS). DD-WRT will run on Buffalo. In addition to Linksys hardware. ASUS. and other access points. and upload it via the router's "firmware update" feature.php/Installation. adjustable transmission power. Linksys has made it known that it does not expect to sell this model in large volume. It includes several useful features.dd-wrt. For specific instructions for your hardware. Specific installation details vary according to the hardware version of your router.com/). . including radio client mode.

Chapter 5: Networking Hardware 155 Figure 5.3: The DD-WRT (v23) control panel .

.

While a certain amount of access control and authentication is necessary in any network. This data can later be used to launch a more sophisticated attack against the network. a user with a high gain antenna may be able to make use of the network from several blocks away. As the network architect. The rules change significantly with wireless networks. While software mechanisms are an important component of network security. limiting physical access to the network devices is the ultimate access control mechanism. is impossible to simply “trace the cable” back to the user s location. Simply put. There s an old saying that the only way to completely secure a computer is to unplug it. Without transmitting a single packet. the network can likely be trusted. even on wired networks. if all terminals and network components are only accessible to trusted individuals. and bury the whole thing in con157 . Disgruntled employees. you have failed in your job if legitimate users find it difficult to use the network to communicate. they can use (or abuse) the network resources.6 Security & Monitoring In a traditional wired network. and simple mistakes on the part of honest users can cause significant harm to network operations. uneducated network users. While the apparent range of your access point may seem to be just a few hundred meters. your goal is to facilitate private communication between legitimate users of the network. It is usually unreasonable to completely trust all users of the network. a nefarious user can even log all network data to disk. access control is very straightforward: If a person has physical access to a computer or network hub. Should an unauthorized user be detected. lock it in a safe. destroy the key. Never assume that radio waves simply “stop” at the edge of your property line.

This avoids the need to smash new holes every time a cable needs to be passed. Where this would be an expensive option in the developed world. outages often occur due to human tampering. Assuring the physical security of an installation is paramount. this type of labour intensive activity can be affordable in Southern countries. Custom enclosures are also easy to manufacture and should be considered essential to any installation. you are building an infrastructure that people depend on.158 Chapter 6: Security & Monitoring crete. PVC can be embedded in cement walls for passing cable from room to room. . Someone may unplug an Ethernet cable so that they can connect their laptop for 5 minutes. hubs or interior access points can be screwed directly onto a wall with a wall plug. In many installations. For many installations. When you make security decisions for your network. such as wires and boxes. Security measures exist to ensure that the network is reliable. or curiosity may lead them to experiment. It is best to put this equipment as high as possible to reduce the chance that someone will touch the device or its cables. or move a switch because it is in their way. or boxes will not be as easy to find. Plastic bags can be stuffed into the conduit around the cables for insulation. it is useless for communication. Small equipment should be mounted on the wall and larger equipment should be put in a closet or in a cabinet. Putting things out of the way and limiting access is the best means to assure that accidents and tinkering do not occur. Switches Switches. Security considerations are important. In less developed economies. While such a system might be completely “secure”. which are easily disturbed. You should be able to find electrical supplies that will work just as well. They may not realize the importance of a cable connected to a port. the network exists so that its users can communicate with each other. people will not understand the purpose of the installed equipment. Networks have physical components. proper fasteners. remember that above all else. A plug might be removed from a power bar because someone needs that receptacle. but should not get in the way of the network s users. Physical security When installing a network. ties. It is often economical to pay a mason to make holes and install conduit. whether accidental or not. Signs and labels will only be useful to those who can read your language.

do not leave any empty receptacles. simple cable attachments can be nailed into the wall to secure the cable. On the UPS and power bar. It is preferable to bury cables. including your UPS is at least 30 cm from the ground. People will have the tendency to use the easiest receptacle. Tape them if necessary. To avoid vermin and insects. it is nicer to keep your server running! Water Protect your equipment from water and moisture. Try choosing well-lit but not prominent places to put equipment. to deter thieves and to keep your equipment safe from winds it is good to overengineer mounts. If that is not possible. or be snagged by a ladder. mount the power bar under a desk. cables should be hidden and fastened. so make these critical ones difficult to use. In moist climates. it is important that the equipment has proper ventilation to assure that moisture can be exhausted. etc. or moisture and heat can degrade or destroy your gear. Consider labeling buried cable with a "call before you dig" sign to avoid future accidental outages. Also try to have a roof over your equipment.Chapter 6: Security & Monitoring 159 Cables At the very least. Any installation on walls should be high enough to require a ladder to reach. Nevertheless. Small closets need to have ventilation. If you cannot find it. to avoid damage from flooding. Masts Equipment installed on a mast is often safe from thieves. Hanging wires might be used for drying clothes. If you do not. The marginal expense will be well worth the trouble. or below the frost level in cold climates. use plastic electrical conduit. so that future cables can be run through the same tubing. rather than to leave them hanging across a yard. pinched or cut. Painting equipment a dull white or grey color reflects the sun and makes it look plain and uninteresting. a strong adhesive tape) to secure the plug into the receptacle. The conduit should be buried about 30 cm deep. though it is nice to have light. Power It is best to have power bars locked in a cabinet. In all cases make sure that your equipment. Panel antennas are often preferred because they are much more subtle and less interesting than dishes. you might find a fan or light plugged into your UPS. It is worth the extra investment of buying larger conduit than is presently required. Also avoid anten- . This will make sure that the cable doesn't hang where it can be snagged. so that water and moisture will not fall onto it. or on the wall and use duct tape (or gaffer tape. It is possible to find plastic cable conduit that can be used in buildings.

Most wireless clients will simply choose any available wireless network when their preferred network is unavailable. as those are items that will attract interest by thieves.4 GHz device. Malicious people may even take advantage of this by setting up access points in strategic locations. it can be very difficult to judge where a wireless user is physically located. . making it possible to compromise the network even further. one can simply tune to the channel being used. voice data. Many network devices are vulnerable to other forms of denial of service attacks as well. it is common for laptop users to accidentally associate to the wrong network. or other 2. where a wifi antenna will be useless to the average thief. in that every computer connected to the network can “see” the traffic of every other user.160 Chapter 6: Security & Monitoring nae that resemble television antennae. it has the unfortunate side effect that denial of service (DoS) attacks are trivially simple. Another serious problem with wireless networks is that its users are relatively anonymous. this problem can be mitigated by the use of encryption. Here are several categories of individuals who may cause problems on a wireless network: • Unintentional users. While it is true that every wireless device includes a unique MAC address that is supplied by the manufacturer. high-gain antennas. such as disassociation flooding and ARP table overflows. As more wireless networks are installed in densely populated areas. By simply turning on a high powered access point. It may also provide passwords and other sensitive data. and log every frame. cordless phone. Multipath effects. Even when the MAC address is known. This data might be directly valuable to an eavesdropper (including data such as email. The user may then make use of this network as usual. these addresses can often be changed with software. completely unaware that they may be transmitting sensitive data on someone else s network. video transmitter. While unlicensed spectrum provides a huge cost savings to the user. As we ll see later in this chapter. To monitor all network traffic on an access point. or online chat logs). Threats to the network One critical difference between Ethernet and wireless is that wireless networks are built on a shared medium. a malicious person could cause significant problems on the network. put the network card into monitor mode. They more closely resemble the old network hubs than modern switches. to try to attract unwitting users and capture their data. and widely varying radio transmitter characteristics can make it impossible to determine if a malicious wireless user is sitting in the next room or is in an apartment building a mile away.

Poorly . These logs are then combined with logs from other war drivers. and are turned into graphical maps depicting the wireless “footprint” of a particular city. eavesdropping is a very difficult problem to deal with on wireless networks. • Eavesdroppers. such as a government or corporate office. and omnidirectional antenna. the user opens the entire network up to potential attacks from the inside. but the data they collect might be of interest to a network cracker. As mentioned earlier. GPS. There are two general classes of rogue access points: those incorrectly installed by legitimate users. an eavesdropper can log all network data from a great distance away. and log or even manipulate all data that passes through it. The vast majority of war drivers likely pose no direct threat to networks. The second class of rogue access point can be very difficult to deal with. or they might find security restrictions on the corporate wireless network too difficult to comply with. users can safely connect to open public networks by using strong encryption.Chapter 6: Security & Monitoring 161 The first step in avoiding this problem is educating your users. The “war driving” phenomenon draws its name from the popular 1983 hacker film. logging the name and location of any networks they find. Arguably. see sites such as http://www. As we will see later in this chapter. this problem is significantly reduced.wifimaps. A malicious person could use this information to illegally access the network there. War drivers are interested in finding the physical location of wireless networks. In the simplest case. such an AP should never have been set up in the first place. and stressing the importance of connecting only to known and trusted networks. They typically drive around with a laptop. • Rogue access points. if your users are trained to use strong encryption.com/ . By installing an inexpensive consumer access point without permission.com/. or to ask permission before joining a new network. While it is possible to scan for unauthorized access points on your wired network. For more information about war driving. a legitimate network user may want better wireless coverage in their office. but war driving makes the problem all the more urgent. For example. Many wireless clients can be configured to only connect to trusted networks. • War drivers. without ever making their presence known. setting a clear policy that prohibits them is very important. and those installed by malicious people who intend to collect data or do harm to the network. “War Games”. it might be obvious that an unprotected access point detected by a war driver is located inside a sensitive building. As we will see later in this chapter. or http://www.nodedb. Again. a malicious person can trick people into using their equipment.netstumbler. http://www. By using a passive monitoring tool (such as Kismet).com/. By installing a high powered AP that uses the same ESSID as an existing network. war drivers who use the popular program NetStumbler can be detected with programs such as Kismet.

while it cannot be completely prevented.com/~chris/driftnet/). Alternately. we can authenticate users based on their MAC address. If you have difficulty convincing others of this problem. proper application of strong encryption will discourage eavesdropping. Maintaining MAC tables on every device can be cumbersome. .162 Chapter 6: Security & Monitoring encrypted data can simply be logged and cracked later. With this feature. When a wireless user tries to associate to the access point. or the association will be denied.org/) or Driftnet (http://www. These tools watch a wireless network for graphical data.etherpeg. While MAC filtering will prevent unintentional users and even most curious individuals from accessing the network. Unfortunately. users should first be authenticated. Later in this chapter. MAC filtering alone cannot prevent attacks from determined attackers. Authentication Before being granted access to network resources. By employing mac filtering on our access points. the access point keeps an internal table of approved MAC addresses. This turns out to be a very difficult problem to solve in the real world. While you can tell a user that their email is vulnerable without encryption. Even worse. these tools simply display all graphics found in a graphical collage. such as GIF and JPEG files. every wireless user would have an identifier that is unique. The closest feature we have to a unique identifier is the MAC address. In an ideal world. you might want to demonstrate tools such as Etherpeg (http://www. unchangeable. we will look at tools and techniques that will help you to mitigate these problems. nothing drives the message home like showing them the pictures they are looking at in their web browser. while unencrypted data can be easily read in real time. MAC addresses can often be changed in software. By observing MAC addresses in use on a wireless network.ex-parrot. This introduction is intended to give you an idea of the problems you are up against when designing a wireless network. and permit all devices that are not on the list. Again. While other users are browsing the Internet. and cannot be impersonated by other users. the AP may keep a table of known “bad” MAC addresses. a determined attacker can spoof (impersonate) an approved MAC address and successfully associate to the AP. the MAC address of the client must be on the approved list. requiring all client devices to have their MAC addresses recorded and uploaded to the APs. This is the 48-bit number assigned by the manufacturer to every wireless and Ethernet device. this is not an ideal security mechanism. I often use tools such as this as a demonstration when lecturing on wireless security.

and use that identity to determine further network access. The most widely employed encryption method on wireless networks is WEP encryption. its MAC address can be added to the filter table to stop the traffic immediately. These frames necessarily contain the network name. the AP does not beacon the ESSID. if a laptop has a virus that sends large amounts of spam or other traffic. and install their own network on the same channel you are using. This will buy you time to track down the user and fix the problem.Chapter 6: Security & Monitoring 163 MAC filters are useful for temporarily limiting access from misbehaving clients. In a typical network. and specifically won t know that you are already using a given channel. WEP uses a shared 40-bit key to encrypt data between the access point and client. But it also means that other network builders cannot easily find your network either. a skilled user can detect frames sent from your legitimate clients to the AP. and is supported by virtually all 802. but the data payload of each packet is encrypted. This will cause interference problems for both you and your neighbor.11a/b/g equipment. With WEP enabled. By using passive monitoring tools (such as Kismet). Another popular authentication feature of wireless the so-called closed network. and users must know the full name of the network before the AP will allow association. wireless clients cannot associate with the AP until they use the correct key. WEP stands for wired equivalent privacy. Encryption also has the benefit of adding a layer of privacy by preventing eavesdroppers from easily watching network traffic. The key must be entered on the APs as well as on each of the clients. see no nearby networks. Through strong encryption. using closed networks ultimately adds little to your overall networks security. Since the network isn t obviously present in site survey tools like NetStumbler. A malicious user can then use this name to associate to the access point. There are a number of drawbacks to this feature. This provides a fairly good authentication mechanism while also adding a bit of privacy to the network. APs will broadcast their ESSID many times per second. A conscientious neighbor may perform a site survey. In a closed network. . we can uniquely identify a user in a manner that is very difficult to spoof. Encryption is probably the best tool we have for authenticating wireless users. just like a normal user would. allowing wireless clients (as well as tools such as NetStumbler) to find the network and display its presence to the user. Finally. An eavesdropper listening to a WEP-enabled network will still see traffic and MAC addresses. This prevents casual users from discovering the network and selecting it in their wireless client. this can prevent your networks from showing up on war driving maps. For example. Forcing users to type in the full ESSID before connecting to the network is error prone and often leads to support calls and complaints.

or if an employee is let go) then changing the password can be prohibitively difficult.html • http://www.berkeley. It provides a significantly stronger encryption scheme.com/papers/others/rc4_ksaproc. the implementation of WEP itself is broken in many access points. For more details about the state of WEP encryption. Anyone who detects the network will see that a key is required. since all APs and client devices need to be changed. WPA was created specifically to deal with the known problems with WEP mentioned earlier. Even worse. and they will likely use other available APs instead. making it clear that they are not welcome to use it.cs. if one user tells a friend what the password is.isaac. This also means that legitimate users of the network can still eavesdrop on each others traffic. all wireless devices support basic WEP. these extensions are not part of the standard.cs. Assuming your users can be trusted not to give away the password. By using WEP on such a link. making it even easier to crack some networks. For one thing. In order to comply with the 802. since they all know the shared key. or WPA. you will discourage others from associating to the link. Think of WEP as a handy “keep out” sign for your network. and generally will not interoperate between equipment from different manufacturers. even on generally open networks. WEP is quite useful for securing long distance point-topoint links.pdf • http://www. If the key is compromised (say. it is certainly the most commonly implemented encryption feature. you can be fairly sure that your wireless clients are legitimate.edu/isaac/wep-faq. you can prevent some of the early attacks found in WEP.164 Chapter 6: Security & Monitoring WEP is definitely not the strongest encryption solution available.ps Another data-link layer authentication protocol is Wi-Fi Protected Access. We will look at other more advanced encryption techniques later in this chapter. While manufacturers have implemented a number of extensions to WEP (such as longer keys and fast rotation schemes).edu/~waa/wireless. it is beyond the skill of most users. By upgrading to the most recent firmware for all of your wireless devices. While it isn t the strongest method available. While WEP cracking is possible. or even SSL certificates to authenticate both the client and the access point. the WEP key is shared between all users.umd. and can use a shared private key.11 standards.1X protocol. see these papers: • http://www. unique keys assigned to each user. WEPs greatest strength is its interoperability. making offline cracking attempts feasible. The key itself is often poorly chosen.crypto. . Authentication credentials are checked using the 802. WEP can still be a useful authentication tool.

To begin. as well as a substantial amount of configuration. They then use their web browser to go to any site on the Internet. WPA requires fairly recent access point hardware and up-to-date firmware on all wireless clients.com/ Internet Captive portal Login: Figure 6. Through the use of Temporal Key Integrity Protocol (TKIP). WPA can be a nightmare to install. Captive portals are typically used on open networks with no other authentication methods (such as WEP or MAC filters). Overall. keys can be rotated quickly over time. Their computer requests a DHCP lease. WPA provides significantly better authentication and privacy than standard WEP.Chapter 6: Security & Monitoring 165 which can consult a third party database such as RADIUS. It can also be used to present information (such as an Acceptable Use Policy) to the user before granting further access. if encryption is used at all. But in most network settings where the vintage of hardware is mixed and the knowledge of wireless users is limited. A captive portal uses a standard web browser to give a wireless user the opportunity to present login credentials. or enter any other credentials that the network administrators require.1: The user requests a web page and is redirected. it solves the rogue access point problem and provides many significant advantages over WEP. This page can require the user to enter a user name and password. The user then . http://google. type in numbers from a pre-paid ticket. further reducing the likelihood that a particular session can be cracked. Captive portals One common authentication tool used on wireless networks is the captive portal. Instead of receiving the requested page. the user is presented with a login screen. By using a web browser instead of a custom program for authentication. a wireless user opens their laptop and selects the network. captive portals work with virtually all laptops and operating systems. WPA can be ideal. If you are installing a network in a setting where you control the entire hardware platform. By authenticating both clients and APs. simply click a “login” button. It is for this reason that most sites continue to use WEP. which is granted.

Since this is not necessarily very secure. or a server anywhere on the Internet.com/ Internet Credentials verified. the user is permitted to access the rest of the network. many implementations will require the user to re-authenticate periodically.166 Chapter 6: Security & Monitoring enters their credentials. Captive portal http://google. instead relying on the MAC and IP address of the client as a unique identifier. All other network access is blocked until these credentials are verified. Once authenticated. captive portals are not a very good choice for networks that need to be locked down to only allow access .com/ Authentication service Figure 6. and is typically redirected to the site they originally requested. Captive portals provide no encryption for the wireless users. another machine on the local network.3: After authenticating. the user is permitted to access network resources. Redirect to http://google. This can often be automatically done by minimizing a special pop-up browser window when the user first logs in. HTTP request waiting Internet User: joe Password: secret Authentication service Captive portal Figure 6. which are checked by the access point or another server on the network. Since they do not provide strong encryption. The authentication server can be the access point itself.2: The user’s credentials are verified before further network access is granted.

Chillispot is a captive portal designed to authenticate against an existing user credentials database. They are much more suited to cafes.chillispot. In public or semi-public network settings. m0n0wall is a complete embedded operating system based on FreeBSD. . • m0n0wall (http://m0n0. requiring them to click a “login” button before using the network. even on untrusted networks such as public access points and the Internet. and other public access locations where casual network users are expected. WiFi Dog provides a very complete captive portal authentication package in very little space (typically under 30kb).net/projects/phpmyprepaid/. encryption techniques such as WEP and WPA are effectively useless. Privacy Most users are blissfully unaware that their private email. users still typically have some expectation of privacy when using computer networks. There is simply no way to distribute public or shared keys to members of the general public without compromising the security of those keys.wifidog. • WiFi Dog (http://www. In these settings. It provides a very easy solution in situations where you need to provide users of an open network with information and an acceptable use policy. pre-paid ticket based authentication can be implemented very easily You can download phpMyPrePaid from http://sourceforge. • NoCatSplash (http://nocat. chat conversations.ch/wall/).Chapter 6: Security & Monitoring 167 from trusted users. a simple application such as a captive portal provides a level of service somewhere between completely open and completely closed. It includes a captive portal with RADIUS support. Privacy can be achieved.info/). such as RADUIS. However mistaken they may be. as well as a PHP web server. allowing it to work on a wider variety of wireless devices. and even passwords are often sent “in the clear” over dozens of untrusted networks before arriving at their ultimate destination on the Internet. The only proven effective method for protecting privacy is the use of strong end-to-end encryption. Popular hotspot projects • Chillispot (http://www. it requires no pop-up or javascript support.org/). From a user s perspective. Combined with the application phpMyPrePaid. This is useful for identifying the operators of the network and displaying rules for network access.net/download/NoCatSplash/) provides a customizable splash page to your users. hotels.

168 Chapter 6: Security & Monitoring Encryption techniques such as WEP and WPA attempt to address the privacy issue at layer two. There is no security in obscurity. This means that legitimate wireless users can eavesdrop on each other. As mentioned earlier. • Data encapsulation. While not an absolute requirement for end-toend encryption. and strong encryption is even stronger when the algorithm is widely known and subject to peer review. Some encryption tools simply provide a secure channel that other applications can use. If the wireless client uses insecure protocols (such as POP or simple SMTP for receiving and sending email). then users beyond the AP can still log the session and see the sensitive data. a user could give sensitive data to anyone claiming to be the legitimate service. Some countries treat encryption as munitions. Before . The encryption algorithm should stand up to public scrutiny. WEP also suffers from the fact that it uses a shared private key. This does protect against eavesdroppers listening in on the wireless connection. where eavesdroppers are listening and possibly even manipulating data coming from the access point. This allows users to run any program they like and still have the protection of strong encryption. Be aware that laws regarding the use of encryption vary widely from place to place. The user should be able to know without a doubt that the remote end is who it claims to be. including DNS lookups and other supporting protocols. By using encryption to the remote end of the connection. and may require a permit. the data-link layer. even if the programs themselves don t support it. users can neatly sidestep the entire problem. even if the key of another user of the service is compromised. good end-to-end encryption should provide the following features: • Verified authentication of the remote end. the use of public key cryptography instead of a shared key can ensure that an individual's data remains private. Without authentication. This can range from encrypting a single email transaction to encapsulation of all IP traffic. It also solves certain problems with distributing keys to users over untrusted networks. or even prohibit its use altogether. but this protection ends at the access point. To ensure data privacy. since they all know the private key. • Public key cryptography. • Strong encryption methods. escrow of private keys. and not be easily decrypted by a third party. A good algorithm with a suitably large and protected key can provide encryption that is unlikely to be broken by any effort in our lifetimes using current technology. A good end-to-end encryption mechanism protects as much data as possible. These techniques work well even on untrusted public networks.

The server optionally verifies the identity of the browser s certificate. we ll take a look at some specific tools that can provide good protection for your users data. SSL The most widely available end-to-end encryption technology is Secure Sockets Layer. the browser and the server first exchange certificates. If the certificates are approved. Built into virtually all web browsers. that it has not expired.4: Eavesdroppers must break strong encryption to monitor traffic over an encrypted tunnel. The SSL implementation built into web browsers includes a collection of certificates from trusted sources. The conversation inside the tunnel is identical to any other unencrypted conversation. Eavesdroppers can watch unencrypted wireless traffic.Chapter 6: Security & Monitoring 169 implementing any solution that involves encryption. you are using SSL. known simply as SSL. Internet Access point Wireless traffic is protected by an encrypted tunnel. SSL uses public key cryptography and a trusted public key infrastructure (PKI) to secure data communications on the web. Whenever you visit a web URL that starts with https. That key is then used to encrypt all communications until the browser disconnects. When you browse to a website that uses SSL. be sure to verify that use of this technology is permitted in your local area. but also prevents so-called man-in-the-middle (MITM) at- . The tunnel can terminate anywhere on the Internet. and that it is signed by a trusted certificate authority. These certificates are cryptographic keys that are used to verify the authenticity of websites. This kind of data encapsulation is known as a tunnel. The browser then verifies that the certificate provided by the server matches its DNS host name. In the following sections. called certificate authorities (CA). The use of certificates with a PKI not only protects the communication from eavesdroppers. the browser and server then negotiate a master session key using the previously exchanged certificates to protect it. Figure 6.

SSL can be used to effectively secure just about any service that runs over TCP. It can use passwords. then such an attack is not possible. SSH Most people think of SSH as a secure replacement for telnet. Unless a CA has been compromised (very unlikely) or the user is tricked into accepting the forged certificate. Instead of a PKI. or even an encrypting web proxy. it uses a key fingerprint cache that is checked before a connection is permitted. especially when using wireless networks. In a man-in-the-middle attack.5: The man-in-the-middle effectively controls everything the user sees. This is why it is vitally important that users understand that ignoring warnings about expired or improper certificates is very dangerous. If your email server does not provide SSL support.stunnel. By first establishing an . Without a public key infrastructure to verify the authenticity of keys. a malicious user intercepts all communication between the browser and the server. the malicious user could carry on two simultaneous encrypted sessions. By presenting counterfeit certificates to both the browser and the server. Use of a good PKI prevents this kind of attack. strong encryption alone cannot protect against this kind of attack. you can still secure it with SSL using a package like Stunnel (http://www. Many people do not know that SSH can also act as a general purpose encrypting tunnel. But SSH is much more than encrypted remote shell. By clicking the “ignore” button when prompted by their browser. In order to be successful. it is trivial to observe and manipulate data passing between the server and the browser. and can record and manipulate all traffic. and SMTP can be secured by wrapping them in an SSL tunnel. SSL is not only used for web browsing. users open themselves up to many potential attacks. the malicious user would have to present a certificate to the client that is signed by a trusted certificate authority. Since the malicious user knows the secret on both connections.org/). public keys. Insecure email protocols such as IMAP. Most modern email clients support IMAPS and POPS (secure IMAP and POP) as well as SSL/TLS protected SMTP. it uses strong public key cryptography to verify the remote server and encrypt data. Like SSL. Server Man-in-the-middle User Figure 6. just as scp and sftp are the secure counterparts of rcp and ftp.170 Chapter 6: Security & Monitoring tacks. or other methods for user authentication. POP.

While this technique may be a bit advanced for many users. insecure protocols can be protected from eavesdropping and attack. suppose you want to forward web proxy traffic over an encrypted link to the squid server at squid. any educated user can implement SSH connections for themselves. These examples will assume that you are using a recent version of OpenSSH. such as wireless point-to-point links.net:3128 squid. OpenSSH (http://openssh.nl/) and WinSCP (http://winscp.putty.net.example.6: The SSH tunnel protects web traffic up to the SSH server itself.org/) is probably the most popular implementation on Unix-like platforms. Internet All traffic sent from SSH server is unencrypted SSH Server SSH listens for a TCP connection on localhost port 3128 All web traffic is encrypted by SSH Web browser uses localhost port 3128 for its proxy Figure 6.net . OpenSSH will also run on Windows under the Cygwin package (http://www. network architects can use SSH to encrypt traffic across untrusted links. providing their own end-to-end encryption without administrator intervention. To establish an encrypted tunnel from a port on the local machine to a port on the remote side. Forward port 3128 (the default proxy port) using this command: ssh -fN -g -L3128:squid.net/) are available for Windows.com/). Free implementations such as Putty (http://www.Chapter 6: Security & Monitoring 171 SSH connection to a trusted location near (or even on) a remote server.example. Since the tools are freely available and run over standard TCP. For example.example.cygwin. use the -L switch.

not just a single TCP port. These are just a few examples of the flexibility of SSH. Note that in order to bind to a local port less than 1024. without the need to set up squid. and away you go. It can even compress the data along the way.example.net The -C switch turns on compression. Being a VPN. This allows you to create an encrypting web proxy. While it can use traditional shared keys. SSH can encrypt data on any TCP port. OpenVPN OpenVPN is a free. Most people find it considerably easier to understand and configure than IPSEC. ssh -fN -D 8080 remote. which can decrease latency on low capacity links. You can add as many port forwarding rules as you like by specifying the -L switch multiple times. Some amount of latency is unavoidable since all encryption/decryption is done in user space. OpenVPN also has some disadvantages. and Solaris. ssh -fNCg -L110:localhost:110 -L25:localhost:25 mailhost. Mac OS X. Windows 2000/XP and higher. By implementing public keys and using the ssh forwarding agent. including ports used for email.example. such as fairly high latency. All web traffic will then be encrypted before transmission to the remote side. including Linux. The -g switch allows other users on your local segment to connect to the local machine and use it for encryption over the untrusted link.172 Chapter 6: Security & Monitoring The -fN switches instruct ssh to fork into the background after connecting. OpenSSH will use a public key for authentication if you have set one up. OpenBSD. Note that this is not a caching proxy. or it will prompt you for your password on the remote side. SSH can also act as a dynamic SOCKS4 or SOCKS5 proxy. NetBSD. and protect your communications with strong encryption and authentication. OpenVPN . it encapsulates all traffic (including DNS and all other protocols) in an encrypted tunnel. you can automate the creation of encrypted tunnels throughout your wireless network. it simply encrypts all traffic. There are OpenVPN client implementations for a wide range of operating systems. You can then configure your web browser to connect to localhost port 3128 as its web proxy service. open source VPN implementation built on SSL encryption. you must have root privileges on the local machine. FreeBSD. but using relatively new computers on either end of the tunnel can minimize this.net Configure your web browser to use SOCKS4 or SOCKS5 on local port 8080.

we recommend this article from Linux Journal: http://www. Privacy and anonymity are important. corporations and individuals. In principle. as well as the source and destination of that data.net/howto. Suppose you want to offer Internet connectivity to your local community by setting up a number of access points for people to connect to. or even down to the Data-Link layer. any one of these routers has the ability to look closely at your data.linuxjournal. VPN technology is a complex field. You can use it to create robust VPN connections between individual machines. OpenVPN needs to connect to a single TCP or UDP port on the remote side. there is always the . your traffic passes through many different routers. It is important to understand how VPNs fit into the structure of your network in order to provide the best possible protection without opening up your organization to unintentional problems. robust encryption protocol (SSL and RSA) • It is relatively easy to configure • It functions across many different platforms • It is well documented • It's free and open source. and is a bit beyond the scope of this section to go into more detail. it can encapsulate all data down to the Networking layer. Often this is enough to piece together a fairly complete picture of your activities on-line. There are many good online resources that deal with installing OpenVPN on a server and client. if your solution requires it. Some of these reasons include: • It is based on a proven.com/article/7949 as well as the official HOWTO: http://openvpn. seeing the source and destination addresses.html Tor & Anonymizers The Internet is basically an open network based on trust. Even if your data is encrypted using a secure protocol. OpenVPN has many advantages that make it a good option for providing end-to-end security. When you connect to a web server across the Internet. and closely linked to each other. Whether you charge them for their access or not. or simply use it to connect network routers over untrusted wireless networks. owned by a great variety of institutions. it is possible for your Internet provider to monitor the amount of data transferred. There are many valid reasons to consider protecting your privacy by anonymizing your network traffic.Chapter 6: Security & Monitoring 173 really shines when used with SSL certificates and a certificate authority. Once established. and quite often also the actual content of the data.

or even a small embedded access point (such as a Linksys WRT54G) where they provides anonymity to all network users automatically.org/) and Privoxy (http://www.com to the appropriate IP address. Network Monitoring Network monitoring is the use of logging and analysis tools to accurately determine traffic flows. the source and destination addresses cannot be linked together. utilization. Mac OSX. Privoxy connects to Tor as a SOCKS4a proxy. Linux and a variety of BSD's. Tor works by repeatedly bouncing your TCP connections across a number of servers spread throughout the Internet. This makes traffic analysis extremely difficult. The need for the Privoxy privacy proxy in connection with Tor is due to the fact that name server queries (DNS queries) in most cases are not passed through the proxy server.com) by the fact that you sent a DNS query to translate google.org/) is a powerful way to run a local proxy server that will pass your Internet traffic through a number of servers all across the net. Combined with secure. The problem is neatly sidestepped if it were technically infeasible to determine where your traffic was actually headed. using Privoxy with Tor is a simple and effective way to prevent traffic analysis from linking your IP address with the services you use online. This means that. and by wrapping routing information in a number of encrypted layers (hence the term onion routing). There are tools that allow you to anonymize your traffic in relatively easy ways. and other performance indicators on a network. In other words. And what about on-line censorship? Publishing web pages anonymously may also be necessary to avoid government censorship.174 Chapter 6: Security & Monitoring risk that people use the network for something that is not legal in your country or region.privoxy. Good monitoring tools give you both hard numbers and graphical ag- . You could plead with the legal system that this particular illegal action was not performed by yourself.torproject. Tor and Privoxy can also be installed on a gateway server. Tor and Privoxy provide a high level of anonymity on the Internet. Tor can be run on a local PC. that get peeled off as the packet moves across the network. but could have been performed by anyone connecting to your network. making it very difficult to follow the trail of information. where it anonymizes traffic from the browser on that particular machine. under Microsoft Windows. and someone analyzing your traffic would easily be able to see that you were trying to reach a specific site (say google. which uses hostnames (not IP addresses) to get your packets to the intended destination. encrypted protocols (such as those we have seen in this chapter). The combination of Tor (http://www. at any given point in the network.

computer performance slows to a crawl (even when not using the network). users begin to complain of slow network speeds and an increase in spam emails. With frequent complaints and very low computer usage. so you know where adjustments may be needed. such as: • What are the most popular services used on the network? • Who are the heaviest network users? • What other wireless channels are in use in my area? • Are users installing wireless access points on my private wired network? • At what time of the day is the network most utilized? • What sites do your users frequent? • Is the amount of inbound or outbound traffic close to our available network capacity? • Are there indications of an unusual network situation that is consuming bandwidth or causing other problems? • Is our Internet Service Provider (ISP) providing the level of service that we are paying for? This should be answered in terms of available bandwidth. the Board is questioning the need for so much network hardware. These tools can help you answer critical questions. The Board also wants evidence that the bandwidth they are paying for is actually being used. It consists of 50 computers and three servers: email. and proxy servers. This helps you to visualize precisely what is happening. And perhaps the most important question of all: • Do the observed traffic patterns fit our expectations? Let's look at how a typical system administrator can make good use of network monitoring tools. packet loss. As the network administrator.Chapter 6: Security & Monitoring 175 gregate representations of the state of the network. latency. causing considerable frustration in your users. How can you diagnose the sudden drop in network and computer performance and also justify the network hardware and bandwidth costs? . web. While initially things are going well. let's assume that we are in charge of a network that has been running for three months. and overall availability. you are on the receiving end of these complaints. An effective network monitoring example For the purposes of example. As time goes on.

The advantages of monitoring external traffic include: • Internet bandwidth costs are justified by showing actual usage. • Network hardware and resources can be justified with real statistics. After installing good anti-virus and anti-spyware software on all of the machines. you should begin by looking at traffic on the local LAN. and the users' morale quickly improves. you need to demonstrate that the bandwidth the organization is paying for is actually what they are getting from their ISP. it becomes obvious that the internal LAN is swamped with far more traffic than the Internet connection can support. SNMP is an application-layer protocol designed to facilitate the exchange of management information between network devices.176 Chapter 6: Security & Monitoring Monitoring the LAN (local traffic) To get an idea of exactly what is causing the slow down. and whether that usage agrees with your ISP's bandwidth charges. With MRTG monitoring in place. as an aggregate average over time. Anything received from (or sent to) a network other than your internal LAN also qualifies as external traffic. so you are able to view the graphs from any machine at anytime. There are several advantages to monitoring local traffic: • Troubleshooting is greatly simplified. . By assigning an IP address to each switch. • Malicious users can be detected and dealt with. observing the entire network from a single point. By using a free tool such as MRTG (see Page 190). External traffic is generally classified as anything sent over a Wide Area Network (WAN). You can achieve this by monitoring external traffic. This is much easier than enabling SNMP on all computers in a network. you can monitor each port on the switch and present data graphically. The graphs are accessible from the web. The machines run much more quickly. Monitoring the WAN (external traffic) In addition to watching the traffic on the internal LAN. spam emails are reduced. • Viruses can be detected and eliminated. Assume that all of the switches support the Simple Network Management Protocol (SNMP). you are able to monitor all the interfaces on that switch. even when the lab is unoccupied. This is a pretty clear indication that some of the computers are infested with a network virus. the internal LAN traffic settles down to expected levels.

Monitoring this traffic is easily done with the use of MRTG on an SNMP enabled device. you now have an accurate measurement of how much bandwidth the organization is using. • Intruders from the Internet are detected and filtered before they can cause problems.7: A graph with a "flat top" is one indication of overutilization. you can make a plan for further optimizing your existing connection (by upgrading your proxy server and using other techniques in this book) and estimate how soon you will need to upgrade your connection to keep up with the demand. and monitor the port traffic just as you would with an internal LAN. and discuss ways to bring actual usage in line with that policy. If your router does not support SNMP. Figure 6. It is clear that your current Internet connection is overutilized at peak times. A "flat top" graph is a fairly clear indication that you are operating at full capacity. Figure 6. then you can add a switch between your router and your ISP connection. It can also indicate the actual throughput of your connection if you are using close to your available capacity at peak times. Detecting Network Outages With monitoring tools in place. causing network lag. . This measurement should agree with your ISP's bandwidth charges. such as a router. This is also an excellent time to review your operational policy with the Board.Chapter 6: Security & Monitoring 177 • Future capacity needs are estimated by watching usage trends and predicting likely growth patterns.7 shows flat tops in peak outbound traffic in the middle of every day except Sunday. After presenting this information to the Board.

is the car any faster or more efficient than it was before? Similarly.178 Chapter 6: Security & Monitoring Later in the week. Good monitoring tools can demonstrate without a doubt that the network infrastructure (bandwidth. With good monitoring tools in place. Network budget and resources are justified. and software) is suitable and able to handle the requirements of network users. You can check the current performance against this history to find unusual behavior. the network comes to life again. Your job is easier. and your users are much happier. When problems do come up. and will alert you to machines that have gone down. alerting you immediately when problems arise. it will send notifications via SMS or email. There are several benefits to implementing a good monitoring system for your network: 1. You are notified automatically when problems arise. rather than waiting for a user to complain? One way to do this is to use a program such as Nagios that continually polls network devices and notifies you of outages. How can you troubleshoot such an outage without such time consuming trial and error? Is it possible to be notified of outages as they occur. how can you pay for an electricity or water bill without seeing your monthly usage from a meter? You must have an account of your network bandwidth utilization in order to justify the cost of services and hardware purchases. and you have historical statistics of how the network devices are performing. hardware. with your eyes closed. Nagios will report on the availability of various machines and services. You then reboot the router. You rush to the lab and hastily reboot the proxy server. Browsing and email are still broken. it is simple to determine the source and nature of the problem. you will be able to justify the cost of equipment and bandwidth by effectively demonstrating how it is being used by the organization. In addition to displaying the network status graphically on a web page. Monitoring your network Managing a network without monitoring is similar to driving a vehicle without a speedometer or a fuel gauge. .a loose power cable is to blame. You continue eliminating the possible fault areas one by one until you realize that the network switch is off . Apparently. no one in the lab can browse the web or send email. and to account for usage trends. you receive an emergency phone call in the evening. After applying power. but there is still no success. with no results. How do you know how fast you are going? Is the car consuming fuel as efficiently as promised by the dealers? If you do an engine overhaul several months later. the Board is satisfied. and head off problems before they become critical.

Fortunately. Network viruses are easily detected. it is impossible to fine tune your devices and protocols to achieve the best possible performance. Without effective monitoring. 7. Proper network usage can be enforced. The dedicated monitoring server While monitoring services can be added to an existing network server. Network intruders are detected and filtered. or if you consume more than 50 Mbps of Internet bandwidth. you will likely need to split up monitoring duties between a few dedicated machines. this makes it possible to build a very capable monitoring server from recycled PC parts. you can detect attackers and prevent access to critical internal servers and services. typically with little CPU power required. network monitoring does not need to be an expensive undertaking. particularly on a busy network. You can be alerted to the presence of network viruses. If you are attempting to account for all services accessed per MAC address. By watching your network traffic. it is often desirable to dedicate one machine (or more. If your network includes more than a few hundred nodes. This depends largely on exactly what you want to monitor.Chapter 6: Security & Monitoring 179 2. The exception to this rule is in very large installations. Some applications (such as ntop) require considerable resources to run. But most logging and monitoring programs have modest RAM and storage requirements. Since open source operating systems (such as Linux or BSD) make very efficient use of hardware resources. and take appropriate action before they consume Internet bandwidth and destabilize your network 4. Some kinds of problems can even be repaired automatically. the only way to be fair to all users is to ensure that the network is being used for its intended purpose. Capacity planning is much easier. this will consume considerably more resources . 3. if necessary) to network monitoring. There are many freely available open source tools that will show you exactly what is happening on your network in considerable detail. 6. Rather than attempting "trial and error" to debug network problems. you do not have to "guess" how much bandwidth you will need as your network grows. With solid historical performance records. There is usually no need to purchase a brand new server to relegate to monitoring duties. you can be instantly notified of specific problems. 5. This section will help you identify many invaluable tools and how best to use them. When bandwidth is a scarce resource. Troubleshooting of network problems is greatly simplified. Network performance can be highly optimized.

If multiple switches are used in an installation. Typically. this means it must have access to the entire network. users. you are effectively blind to problems with the network. That connection can either be a physical cable. To a network administrator. To monitor a WAN connection. Figure 6. and that web server develops problems. the monitoring server may need a connection to all of them. While the inbound and outbound utilization are clear. or if . While consolidating monitoring services to a single machine will streamline administration and upgrades. it can also ensure better ongoing monitoring. To monitor a LAN. This provides simple feedback about utilization. you can do this from just about anywhere on the LAN.8 shows a typical MRTG graph generated from the Internet router. Your monitoring should be robust and protected from service outages as well as possible. a single dedicated monitoring machine is usually enough.180 Chapter 6: Security & Monitoring than simply measuring network flows on a switch port. but you cannot break the data down further into machines. services. Where does the server fit in my network? If you are only interested in collecting network flow statistics from a router. But for the majority of installations. but cannot give you comprehensive details about usage patterns. there is no detail about which computers. such as the Internet link to your ISP. the monitoring server is typically connected to a monitor port on the switch. the monitoring server must be able to see the traffic passing through the edge router. Without network statistics. if you install monitoring services on a web server.8: Polling the edge router can show you the overall network utilization. the dedicated monitoring server must have access to everything that needs to be watched. and users. Figure 6. or protocols are using bandwidth. For more detail. For example. the data collected about network performance is nearly as important as the network itself. then your network may not be monitored until the problem is resolved.

it introduces a single point of failure for the network.10: By inserting a network monitor between the LAN and your Internet connection.9: Use the monitor port on your switch to observe traffic crossing all of the network ports. the monitoring server may be installed between your internal LAN and the Internet. if the server cannot keep up with the demands of the network. It is also a potential performance bottleneck. Internet Network monitor Figure 6. a VLAN specifically configured for monitoring traffic. . If monitor port functionality is not available on your switch. you can observe all network traffic.Chapter 6: Security & Monitoring 181 your network switches support it. While this will work. as the network will fail if the monitoring server develops a problem. Internet Switch Monitor port Network monitor Figure 6.

hubs are generally considered to be much more reliable than routers. While this does still introduce an additional point of failure to the network (since the entire network will be unreachable if the hub dies). and connect the monitoring server to the hub. Internet Hub Network monitor Figure 6. external router. They are also very easily replaced should they fail. you can insert a network hub between your Internet router and the LAN. What to monitor It is possible to plot just about any network event and watch its value on a graph over time.11: If your switch does not provide monitor port functionality.182 Chapter 6: Security & Monitoring A better solution is to use a simple network hub (not a switch) which connects the monitoring machine to the internal LAN. and the monitoring machine. you will have to decide what information is important in order to gauge the performance of your network. Since every network is slightly different. Wireless statistics • Received signal and noise from all backbone nodes • Number of associated stations • Detected adjacent networks and channels • Excessive retransmissions . Here are some important indicators that many network administrators will typically track. Once your monitoring server is in place. you are ready to start collecting data.

As your network matures. fan speed. and system voltages • Disk SMART status • RAID array status You should use this list as a suggestion of where to begin. and you should of course track those as well. email servers.). • Ping times and packet loss rates to your ISP • Status of backups System health statistics • Memory usage • Swap file usage • Process count / zombie processes • System load • Uninterruptible Power Supply (UPS) voltage and load • Temperature. There are many freely available . you will likely find new key indicators of network performance.Chapter 6: Security & Monitoring 183 • Radio data rate. etc. if using automatic rate scaling Switch statistics • Bandwidth usage per switch port • Bandwidth usage broken down by protocol • Bandwidth usage broken down by MAC address • Broadcasts as a percentage of total packets • Packet loss and error rate Internet statistics • Internet bandwidth use by host and protocol • Proxy server cache hits • Top 100 sites accessed • DNS requests • Number of inbound emails / spam emails / email bounces • Outbound email queue size • Availability of critical services (web servers.

Intrusion detection tools watch for undesirable or unexpected network traffic. Network detection tools listen for the beacons sent by wireless access points. They let you quickly detect nearby networks and determine if they are in range or are causing interference. protocol information. and channel. and take appropriate action (typically denying access and/or notifying a network administrator). and provisioning extra capacity before you run out. • The built-in client. You can predict and avoid such problems by monitoring the number of available modems. or if any are faulty. You should consider monitoring the availability of any resource where unavailability would adversely affect your network users. then users will be denied access and will probably complain. A monitoring machine that is low on resources can affect your ability to monitor the network effectively. which inspect every packet on the network and provide complete detail about any network conversation (including source and destination addresses. Throughput testing tools tell you the actual bandwidth available between two points on a network. Network detection The simplest wireless monitoring tools simply provide a list of available networks. and even application data). While . Realtime monitoring tools perform similar monitoring. along with basic information (such as signal strength and channel). This typically includes the ability to scan for available networks. All modern operating systems provide built-in support for wireless networking. and display information such as the network name. If all the modems are used. received signal strength. benchmarking tools estimate the maximum performance of a service or network connection. but notify administrators immediately if they detect a problem. since it generates traffic by polling a particular machine. for example its CPU usage and disk space. Types of monitoring tools We will now look at several different classes of monitoring tools.184 Chapter 6: Security & Monitoring tools that will show you as much detail as you like about what is happening on your network. Spot check tools are designed for troubleshooting and normally run interactively for short periods of time. allowing the user to choose a network from a list. For example. and typically plot the results on a graph. Passive spot check tools include protocol analyzers. Don't forget to monitor the monitoring machine itself. A program such as ping may be considered an active spot check tool. in order to receive advance warning if it becomes overloaded or faulty. Trending tools perform unattended monitoring over long periods. your users may dial into modems on your site to gain remote access to your network. Finally.

then you ll need to be able to isolate the exact location of the problem. From the makers of • Ministumbler (http://www. Spot check tools What do you do when the network breaks? If you can t access a web page or email server.macstumbler. Troubleshooting. functionality can vary widely between implementations. and is very easy to use. It uses ICMP packets to attempt to contact a specified host. and clicking the reload button doesn t fix the problem. Wellenreiter is a nice graphical wireless network detector for Linux. They tend to provide little information apart from network names and the available signal to the access point currently in use. It also features a signal/noise meter that plots radio receiver data as a graph over time. Macstumbler provides much of the same functionality but for the Mac OS X platform. For more discussion of common network problems and how to diagnose them. Ministumbler is handy to run on a handheld PDA with a wireless card for detecting access points in the field.netstumbler. .com/). but cannot detect “closed” wireless networks. It also integrates with a variety of GPS devices. It requires Perl and GTK. • Wellenreiter (http://www.com/). This makes Netstumbler a handy tool to have for an informal site survey. It supports a variety of wireless cards.wellenreiter.Chapter 6: Security & Monitoring 185 virtually all wireless devices are guaranteed to have a simple scanning utility. and of course Linux and BSD) includes a version of the ping utility. While not directly related to the Netstumbler. but works on the Pocket PC platform. Ministumbler provides much of the same functionality as the Windows version. see Chapter 9. This is the most popular tool for detecting wireless networks using Microsoft Windows.netstumbler. It works with all Apple Airport cards. and tells you how long it takes to get a response. • Macstumbler (http://www.net/). Lucent. ping Just about every operating system (including Windows.com/). Netstumbler. for logging precise location and signal strength information. and supports Prism2. Mac OS X. These tools will help you to determine just where a connection problem exists. This section is simply an introduction to commonly used troubleshooting tools. • Netstumbler (http://www. These tools are typically only useful for configuring a computer in a home or office setting. and Cisco wireless cards. It will detect open and encrypted networks.

94.186 Chapter 6: Security & Monitoring Knowing what to ping is just as important as knowing how to ping.90.13): 56 data bytes 64 bytes from 66.235. If you find that you cannot connect to a particular service in your web browser (say.90.13: icmp_seq=2 ttl=56 time=34. If you can t even ping other IP addresses on your local LAN.94.230 (69. is it plugged in? If you re using wireless.216.bitwizard. traceroute and mtr http://www.38. If return ping packets have an unusually low Time To Live (TTL).991 ms 64 bytes from 69.375 ms 64 bytes from 66. http://yahoo. then chances are you won t be able to get to the Internet either. As with ping. Since you will likely find ping on just about any machine you will work on.231.90.com/). and is it in range? Network debugging with ping is a bit of an art.com ping statistics --3 packets transmitted.230: icmp_seq=0 ttl=126 time=12.467/2. If you can t reach it.90.235.1 ping statistics --3 packets transmitted.yahoo. 0% packet loss round-trip min/avg/max/stddev = 12.869 ms 64 bytes from 69. it s a good idea to learn how to use it well. it s a good idea to see if you can ping your default router: $ ping 69.13: icmp_seq=0 ttl=57 time=29.90. Try pinging an IP address on the Internet. But what if the ping doesn t return any data at all? If you are pinging a name instead of an IP address.618 ms Hit control-C when you are finished collecting data.234. are you connected to the proper wireless network. If packets take a long time to come back. then it s time to check your connection. By running traceroute.000/35. you could try to ping it: $ ping yahoo. you may be running into DNS problems.897 ms ^C --. 3 packets received.235. you can find the location of problems between your computer and any point on the Internet: . you may have routing problems between your machine and the remote end.991/13.869/0.230: icmp_seq=1 ttl=126 time=14.234.767 ms If you can t ping your default router.235. If you re using Ethernet.nl/mtr/.375/33.919/14. there may be network congestion. traceroute is found on most operating systems (it s called tracert in some versions of Microsoft Windows).234. but it is useful to learn.230): 56 data bytes 64 bytes from 69.230: icmp_seq=2 ttl=126 time=13.158 ms ^C --.94.230 PING 69.com (66.235.235.234.90. 0% packet loss round-trip min/avg/max/stddev = 29.13: icmp_seq=1 ttl=56 time=35.467 ms 64 bytes from 66.com PING yahoo.94. 3 packets received.

14.sfo1.0% 4 89.0% 4 91. For wired networks.yahoo. er1.32. For wireless networks.207.0. Here are several popular (and free) network protocol analyzers: .0 1.pat1.8 2328.2 4.788 ms 28.0% 4 11.763 ms 1.5 1.1 36.com 0.8 99. it might be worth starting your troubleshooting effort there.0% 3 91.17.yahoo.9 29.ge-0-1-0. you can inspect information all the way down to individual 802. bas1-m.1 8.809 ms 2528.5 14. Protocol analyzers Network protocol analyzers provide a great deal of detail about information flowing through a network. By running mtr.0 12.7 28. the round trip time shoots up to more than two seconds.Chapter 6: Security & Monitoring 187 $ traceroute -n google.rob. by allowing you to inspect individual packets. fe-0-3-0.4 2.4 7.731 ms 2 216.0 89.7 10. 40 byte packets 1 10.3 93.speakeasy.dcn. and makes the trace run more quickly.7 91.115 ms 6 66.cr2.719 ms 8 * * * The -n switch tells traceroute not to bother resolving names in DNS.2 93. w2.1 4.dcn.speakeasy.0% 4 27.3 3.7 38.322 ms 1.1 The data will be continuously updated and averaged over time.2 4. while packets seem to be discarded at hop eight.p400.15.vip.11 frames.com traceroute to google.6 27. tesla.478 ms 29.net 0.7 93.601 ms 29.94.4 6.0 89.yahoo.7 15. so-1-1-0.yahoo.speakeasy. Note that you must have root privileges to run mtr. You can see that at hop seven. 0. ge5-2.4 1. While these tools will not revel precisely what is wrong with the network. you can get an ongoing average of latency and packet loss to a single host.256 ms 13.14 28.69] (tos=0x0 psize=64 bitpatSun Jan 8 20:01:26 2006 Restart statistics Order of fields quit Packets Pings Host Loss% Snt Last Avg Best Wrst StDev 1.176.231.0% 4 1.net 0.com 0.561 ms 3 69.7 2. ae1. instead of the momentary snapshot that ping and traceroute provide. As with ping.0% 4 89.811 ms 7 172.1 5.16.com (72.1 90.pao. they can give you enough information to know where to continue troubleshooting.38.0 2.sea1. you can inspect packets at the data-link layer or above.913 ms 28.6 4.0.31 40.swn 0. If this part of the network is in your control.0% 4 36.648 ms 13.msr1.249.545 ms 27. This might indicate a problem at that point in the network.cr2.8 99.236.99).83. 220.swn (0.0 1.bas1-m.160 ms 28.1 90.9 33. My TraceRoute (mtr) is a handy program that combines ping and traceroute into a single tool.83.yahoo.197 ms 13.17.sea1.9 9.rob.0% 4 15.dcn.233 14.0 1.rc. you should hit control-C when you are finished looking at the data.7 14.267 ms 4 69. gremlin.2 93.494 ms 5 198.150 32.0) Keys: Help Display mode My traceroute [v0.6.3 91.0 34.6 0.0 11.9 2.944 ms 2428.com 0.dce.1 4. 64 hops max.com 0.com 0.187 ms 14.

ch/. but with a slick Mac OS X graphical interface. Kismet is a powerful wireless protocol analyzer for many platforms including Linux. and have them all report over the network back to a central user interface. it can even detect “closed” wireless networks by analyzing traffic sent by wireless clients.188 Chapter 6: Security & Monitoring Kismet http://www. for later analysis with tools like Ethereal.net/. Kismet is an invaluable tool for diagnosing wireless network problems.macpirate. Exclusively for the Mac OS X platform. KisMAC http://kismac. KisMAC does much of what Kismet can do. and even the embedded OpenWRT Linux distribution.12: Kismet running on a Nokia 770 Internet Tablet Since Kismet uses the radio card's passive monitor mode. It is a passive scanner that will log data to disk in PCAP format compatible with Wireshark. You can run Kismet on several machines at once. AP hardware fingerprinting. such as a university or corporate campus.kismetwireless. and GPS integration. In addition to basic network detection. Kismet also features associated client information. Kismet will passively log all 802. . Mac OS X. It works with any wireless card that supports passive monitor mode.11 frames to disk or to the network in standard PCAP format. Since it is a passive network monitor. Figure 6. it does all of this without transmitting any data. Netstumbler detection. This allows for wireless monitoring over a large area. It supports passive scanning with AirportExtreme cards as well as a variety of USB wireless adapters.

org/windump/. including . and Mac OS X).Chapter 6: Security & Monitoring 189 tcpdump http://www. Wireshark is a free network protocol analyzer for Unix and Windows. There is also a Windows port called WinDump available at http://www. Wireshark http://www. Packets captured with tcpdump can be loaded into wireshark for visual analysis and further diagnostics. It can show all of the packet headers and data received.tcpdump. Wireshark allows you to examine data from a live network or from a capture file on disk. tcpdump is a command-line tool for monitoring network traffic.org/. Tcpdump can capture and display all network protocol information down to the link layer." Figure 6. It is billed as "The World's Most Popular Network Protocol Analyzer. and interactively browse and sort the captured data. Formerly known as Ethereal. The tcpdump tool is available as a standard tool in Unix derivatives (Linux.13: Wireshark (formerly Ethereal) is a powerful network protocol analyzer that can show you as much detail as you like about any packet. It does not have all the bells and whistles of wireshark but it does use fewer resources. Both summary and detailed information is available for each packet.winpcap.wireshark. This is very useful if you wish to monitor an interface on a remote system and bring the file back to your local machine for analysis. BSD. or just the packets that match particular criteria.org/.

MRTG can be a little confusing to set up. These are typically displayed on a web page. MRTG http://oss. RRDtool http://oss. Trending tools collect data as well as analyze and report on it.oetiker. but it can be also used as a general purpose fault finding tool. especially if you are not familiar with SNMP. Trending tools Trending tools are used to see how your network is used over a long period of time. a machine infected with a network worm or virus can be identified by looking for the machine that is send out the same sort of TCPIP packets to large groups of IP addresses. and displaying a summary in a human-readable form (such as a graph). including a rich display filter language and the ability to view the reconstructed stream of a TCP session. as they are not stand-alone programs. Below are some examples of trending tools. For example. It is typically used to isolate and analyze specific traffic to or from an IP address. unless you change something on the system that is being monitored (such as its IP address). The Multi Router Traffic Grapher (MRTG) monitors the traffic load on network links using SNMP.oetiker. MRTG generates graphs that provide a visual representation of inbound and outbound traffic. Figure 6. They work by periodically monitoring your network activity. RRD is short for Round Robin Database. Wireshark has several powerful features. RRD is a database that stores information in a very compact way that does not .ch/rrdtool/.ch/mrtg/. But once it is installed.14: MRTG is probably the most widely installed network flow grapher.190 Chapter 6: Security & Monitoring the full header and data portions. It can be daunting to use for first time users or those that are not familiar with the OSI layers. Some of them need to be used in conjunction with each other. MRTG requires virtually no maintenance.

protocol. as well as generate useful graphs to present the data. ntop http://www. For historical traffic analysis and usage. destination. ntop can use a lot of CPU and disk space. etc. and Windows. or server load average) and can display that data as an average over time.Chapter 6: Security & Monitoring 191 expand over time. It runs on Linux. you will certainly want to investigate ntop.ntop. • Traffic statistics grouped by protocol and port number • An IP traffic matrix which shows connections between machines • Network flows for routers or switches that support the NetFlow protocol • Host operating system identification . It integrates with rrdtool. and can be downloaded from http://oss. It is used to keep track of time-series data (such as network bandwidth. RRDtool is included in virtually all modern Linux distributions. RRDtool refers to a suite of tools that allow you to create and modify RRD databases. This program builds a detailed real-time report on observed network traffic. Mac OS X.). machine room temperature. Note that RRDtool itself does not contact network devices to retrieve data. BSD. Some of its more useful features include: • Traffic display can be sorted by various criteria (source.ch/rrdtool/. but it gives you extensive insight into how your network is being used. MAC address. On very busy networks.org/. It is merely a database manipulation tool. and makes graphs and charts visually depicting how the network is being used. displayed in your web browser. Figure 6.oetiker. RRDtool is also used by many full featured front-ends that present you with a friendly web interface for configuration and display.15: RRDtool gives you a lot of flexibility in how your collected network data may be displayed. RRD graphs give you more control over display options and the number of items available on a graph as compared to MRTG. You can use a simple wrapper script (typically in shell or Perl) to do that work for you.

The main disadvantage of ntop is that it does not provide instantaneous information. This can make it difficult to use to diagnose a problem that starts suddenly. and Ubuntu. including RedHat. depending on the amount of traffic observed.net/. . If you are going to run it for long periods you should monitor the CPU utilization of the monitoring machine. It stores all of the necessary information to create graphs in a MySQL database. ntop can be fairly CPU intensive. PHP. Cacti http://www. Debian. The front-end is written in PHP.org/ and is available for most operating systems. and Python API Ntop is available from http://www. While it can be left running to collect historical data. data sources. It is often included in many of the popular Linux distributions. Cacti does the work of maintaining graphs. only long-term totals and averages.ntop.cacti.192 Chapter 6: Security & Monitoring • P2P traffic identification • Numerous graphical charts • Perl.16: ntop displays a wealth of information about how your network is utilized by various clients and servers. Cacti is a front-end for RRDtool. Figure 6.

There are hundreds of templates for various systems available on the cacti website. NetFlow provides valuable information about network users and applications. peak usage times. network planning. There is support for SNMP devices.17: Cacti can manage the polling of your network devices. NetFlow NetFlow is a protocol for collecting IP traffic information invented by Cisco. Figure 6. usage-based network billing. It also provides more granular information than SNMP. letting you get a more detailed picture of port and protocol usage. NetFlow is also less CPU-intensive on Cisco routers than using SNMP. and traffic routing.Chapter 6: Security & Monitoring 193 and handles the actual data gathering. and the code is under rapid development. and can build very complex and informative visualizations of network behavior. Cacti can be somewhat confusing to configure. Denial of Service monitoring capabilities. security. and network monitoring. Cisco routers can generate NetFlow information which is available from the router in the form of UDP packets. From the Cisco website: Cisco IOS NetFlow efficiently provides a key set of services for IP applications. it can yield very impressive graphs. but once you work through the documentation and examples. and custom scripts can easily be written to poll virtually any conceivable network event. . including network traffic accounting.

To be notified when network problems occur. There are several commercial and free NetFlow collectors available. plain text or a graphical format.18: A typical flow chart generated by Flowc. By analyzing flow data. Flowc uses a MySQL database to store aggregated traffic information. Therefore. latency distribution and packet loss all on a single graph. For more information about NetFlow. The large gap in data probably indicates a network outage. It is lightweight and easy to configure.ch/smokeping/. SmokePing uses RRDtool for data storage. see http://en.org/wiki/Netflow .wikipedia.194 Chapter 6: Security & Monitoring This information is collected by a NetFlow collector that stores and presents the data as an aggregate over time.oetiker. Figure 6. by just looking at a quick snapshot of data during a network crisis. It can also be desirable to use Netflow as a spot check tool. Trending tools typically will not notify you of outages. Ntop is one free tool that can act as a NetFlow collector and probe. or use the included report generators.ua/flowc/. but merely log the occurrence. Think of NetFlow as an alternative to SNMP for Cisco devices. The built-in report generators produce reports in HTML. and can draw very informative graphs that present up to the minute information on the state of your network connection. SmokePing http://oss. use a realtime monitoring tool such as Nagios (see Page 200).kiev. store and display latency. . It can measure. Flowc is an open source NetFlow collector (see NetFlow above). it is possible to generate your own reports from the data using SQL. Flowc http://netacad. one can build a picture of traffic flow and traffic volume in a network or on a connection. Another is Flowc (see below). SmokePing is a deluxe latency measurement tool written in Perl.

but its resource requirements are much lighter. IP checksum errors. It is a simple to use program that uses minimal system resources. iptraf http://iptraf. It has an ncurses interface and runs in a command shell.Chapter 6: Security & Monitoring 195 It is very useful to run SmokePing on a host with good connectivity to your entire network.19. EtherApe http://etherape.sourceforge.net/. EtherApe doesn't show quite as much detail as ntop. Hosts and links change size depending on the amount of traffic sent and received. SmokePing can optionally send alerts when certain conditions are met. . you can observe the effect that network congestion has on packet loss and latency. IPTraf is a lightweight but powerful LAN monitor. Combined with MRTG (see Page 190) or Cacti (see Page 192). Figure 6. Over time. While it does not keep historical data. trends are revealed that can point to all sorts of network problems. As with wireshark and tcpdump. ICMP and OSPF information. EtherApe displays a graphical representation of network traffic.19: SmokePing can simultaneously display packet loss and latency spreads in a single graph. An example of SmokePing in action is shown in Figure 6. such as when excessive packet loss is seen on a link for an extended period of time. IPTraf takes a moment to measure observed traffic. and more. it is very useful for displaying an instantaneous usage report. data can be captured "off the wire" from a live network connection or read from a tcpdump capture file. and then displays various network statistics including TCP and UDP connections. The colors change to represent the protocol most used.seul. traffic flows.org/.

196 Chapter 6: Security & Monitoring Figure 6. Argus is also the name of the mythological Greek god who had hundreds of eyes. and most other UNIX systems. Argus can be deployed to monitor individual end-systems. delay. to allow flexible strategies for collecting network audit data. capacity. generating an audit log of all the network activity seen in the packet stream. demand. Argus data clients support a range of operations. examining data from a live interface. Linux. Argus runs on BSD. such as sorting. Argus provides both push and pull data handling models. Argus http://qosient. As a continuous monitor. From the Argus website: Argus generates flow statistics such as connectivity. loss. aggregation.20: iptraf's statistical breakdown of traffic by port.com/argus/. Argus consists of two parts: a master collector that reads packets from a network device. and a client that connects to the master and displays the usage statistics. or an entire enterprises network activity. and jitter on a per transaction basis. archival and reporting. Argus can be used to analyze and report on the contents of packet capture files or it can run as a continuous monitor. . Argus stands for Audit Record Generation and Utilization System.

dslreports. Here are a few tools that will allow you to perform throughput testing on your own networks. NeTraMet runs on DOS and most UNIX systems. but only the speed of your link to a particular site on the Internet. NeTraMet is another popular flow analysis tool. but don't always give you an accurate picture of network performance.com/stest or http://speedtest. these tests are increasingly inaccurate as you get further from the testing source. and can include Ethernet.Chapter 6: Security & Monitoring 197 NeTraMet http://freshmeat. or other identifiers. Figure 6. .net are pretty. Flows are specified using a simple programming language that define the addresses used on either end. Even worse. Throughput testing How fast can the network go? What is the actual usable capacity of a particular network link? You can get a very good estimate of your throughput capacity by flooding the link with traffic and measuring how long it takes to transfer the data. protocol information.net/projects/netramet/. Like Argus. they do not allow you to test the speed of a given link. While there are web pages available that will perform a “speed test” in your browser (such as http://www. NeTraMet consists of two parts: a collector that gathers statistics via SNMP. including Linux and BSD.21: Tools such as this one from SpeedTest.net/). and a manager that specifies which flows should be watched. IP.

Multiply the result by 8 to find the speed in kilobits per second.76 KB/sec +++ 2048 I/O calls.1 port 1212 connected with 10. . launch a server on one side and a client on the other: node_a$ iperf -s node_b$ iperf -c node_a -----------------------------------------------------------Client connecting to node_a. ttcp is a simple network performance testing tool. It supports many of the same features as ttcp.14 real seconds = 65.57. iperf is a commandline tool for estimating the throughput of a network connection.15.3 sec 768 KBytes 558 Kbits/sec The server side will continue to listen and accept client connections on port 5001 until you hit control-C to kill it.mil/ftp/pub/ttcp/.nlanr. The first node runs in receive mode. not kilobits. Remember that the speed readout is in kilobytes. Fortunately. calls/sec = 8. you should reverse the transmit and receive partners to test the link in the other direction.6. align=16384/0.15. msec/call = 124. and the other transmits: node_a$ ttcp -r -s node_b$ ttcp-t: ttcp-t: ttcp-t: ttcp-t: ttcp-t: ttcp-t: ttcp -t -s node_a buflen=8192.2sys 4:09real 0% 0i+0d 0maxrss 0+0pf 7533+0csw After collecting data in one direction. the code has been released in the public domain and is freely available. ttcp is found as a standard tool on many systems.0user 0. To run iperf.0-11. One instance is run on either side of the link you want to test. iperf http://dast. but uses a “client” and “server” model instead of a “receive” and “transmit” pair.198 Chapter 6: Security & Monitoring ttcp http://ftp. Much like ttcp.6.arl. Like ping and traceroute.22 0. and can alter various TCP parameters and buffer lengths to give the network a good workout. Now a standard part of most Unix-like systems.0 KByte (default) -----------------------------------------------------------[ 5] local 10. nbuf=2048.net/Projects/Iperf/. It can test UDP as well as TCP streams. port=5001 tcp -> node_a socket connect 16777216 bytes in 249. The only real disadvantage to ttcp is that it hasn t been developed in years. It can even use a user-supplied data stream instead of sending random data.23 port 5001 [ ID] Interval Transfer Bandwidth [ 5] 0. This can make it handy when running multiple test runs from a variety of locations. TCP port 5001 TCP window size: 16.

Chapter 6: Security & Monitoring 199 The biggest difference between ttcp and iperf is that iperf is under active development. and packet matching. and many other kinds of anomalous traffic patterns. and has many new features (including IPv6 support). such as stealth port scans. While it is not always as accurate as a flood test. It can be used to detect a variety of attacks and probes. Realtime tools It is desirable to find out when people are trying to break into your network. and depending on the amount of network traffic. It features rulebased logging and can perform protocol analysis. bing can give you a rough idea of network performance without running up the charges that a flood test would certainly incur.fr/bing/index-en. Installing and running Snort is not trivial. so it can estimate available bandwidth without the need to run a special client on the other end. Bing attempts to estimate the available throughput of a point-to-point connection by analyzing round trip times for various sized ICMP packets. Snort is very well documented and has a strong user community. SMB probes. . you can identify unexpected behavior that would otherwise mysteriously eat up your Internet bandwidth.shtml. content searching. bing http://fgouget. The following are some open source tools that can help perform this task. Since it uses relatively little bandwidth. will likely require a dedicated monitoring machine with considerable resources. or when some part of the network has failed. Fortunately.org/) is a packet sniffer and logger which can be used as a lightweight network intrusion detection system. This makes it a good choice as a performance tool when building new networks. Snort Snort (http://www. it can provide a good estimate without transmitting a large number of bytes. there are programs that constantly monitor the status of the network and can send alerts when notable events occur. Since bing works using standard ICMP echo requests. CGI attacks. OS fingerprinting attempts.snort.free. Because no system administrator can be monitoring a network all the time. Snort has a realtime alert capability that can notify administrators about problems as they occur with a variety of methods. and can even attempt to estimate the throughput of links outside your network. Rather than flood a connection with data and see how long the transfer takes to complete. By implementing a comprehensive Snort rule set.

and checks the results against an expected response. ModSecurity increases web application security by protecting web applications from known and unknown attacks. For example. and provides a web interface to show up-tothe-minute system status. Nagios Nagios (http://nagios. which maintains a huge and frequently updated repository of rules: http://gotroot. but Nagios can actually retrieve a web page or make a database request. It performs checks by running small scripts at regular intervals. or in the server being used to launch attacks or send spam to other Internet users.org/). such intrusions can seriously reduce your available bandwidth.apache. SMS. This kind of security tool is also known as a web application firewall. There are several sources for updated mod_security rules that help protect against the latest security exploits. It can send notifications via email. This can yield much more sophisticated checks than a simple network probe.modsecurity.php?page=mod_security+rules Web application security is important in defending against attacks on your web server. As well as being damaging to the Internet as a whole.org/) is a program that monitors hosts and services on your network. or by running a script. Nagios runs on Linux or BSD. ping (page 185) may tell you that a machine is up.org/docs/ for an extensive list of installation and configuration resources.org/) is an open source intrusion detection and prevention engine for web applications. Nagios is extensible.com/tiki-index. or as a module in the Apache web server (http://www.200 Chapter 6: Security & Monitoring See http://snort. and can monitor the status of virtually any network event. and nmap may report that a TCP port responds to requests. It can be used on its own. notifying you immediately when problems arise. which could result in the theft of valuable or personal data. . One excellent resource is GotRoot. Apache: mod_security ModSecurity (http://www. and verify that the response is not an error. and will send notifications to the relevant person or group depending on the nature of the problem.

Chapter 6: Security & Monitoring 201 Figure 6. machine room temperature. This can give you advance warning of network problems. and performs all of the functions you would expect from a modern realtime monitor (such as SNMP polling and instant notification of error conditions).org/) is an open source realtime monitoring tool that is something of a hybrid between Cacti and Nagios.zabbix. or other network health indicator crosses a particular threshold. Driftnet and Etherpeg. Other useful tools There are thousands of free network monitoring tools that fill very specialized needs. These tools decode graphical data (such as GIF and JPEG files) and display them as a collage.22: Nagios keeps you informed the moment a network fault or service outage occurs. It uses a SQL database for data storage. As mentioned earlier. Nagios can even notify you when bandwidth usage. Zabbix is released under the GNU General Public License. Here are a few of our favorites that don't quite fit into the above categories. packet loss. tools such as these are of limited use in . often allowing you to respond to the problem before users have a chance to complain. Zabbix Zabbix (http://www. has its own graph rendering package.

It currently recognizes IPv4 and IPv6.org/.etherpeg. UDP. ngrep Ngrep provides most of GNU grep's pattern matching features. Token Ring. As it makes extensive use of regular expression matches. or other criteria using BPF filters. to view all packets that contain the string GET (presumably HTTP requests). To view GET or POST strings sent to destination port 80.202 Chapter 6: Security & Monitoring troubleshooting problems. Figure 6. This is the filter language used by common packet sniffing tools. and Driftnet can be downloaded at http://www. and much more. IGMP. ports. use this command line: # ngrep -q 'GET|POST' port 80 . it is a tool suited to advanced users or those that have a good knowledge of regular expressions.ex-parrot. For example. but are very valuable for demonstrating the insecurity of unencrypted protocols. but applies them to network traffic. such as tcpdump and snoop. Etherpeg is available from http://www. But you don't necessarily need to be a regex expert to be able to make basic use of ngrep. TCP.com/~chris/driftnet/. try this: # ngrep -q GET Pattern matches can be constrained further to match particular protocols. SLIP.23: A web collage generated by Etherpeg. ICMP. PPP. FDDI.

routing loops. but given some work you can determine what is normal for your network.). Fortunately. For example. which may not indicate a problem. only that it was relatively high. caches. and so on) This is not a definitive list. • The competence of your computer users • The location and configuration of control structures (firewalls. quotas. You can download ngrep at http://ngrep. types. it may mean that your network has been infected with a virus. proxy servers. and level of services offered • The health of the network (presence of viruses. or a potential future problem. With this in mind. .). open email relays. with your network. What is normal? If you are looking for a definitive answer as to what your traffic patterns should look like. you need to determine for yourself what your traffic patterns look like under normal situations. and you are not sure of the cause.Chapter 6: Security & Monitoring 203 By using ngrep creatively. you have decided to keep a graph of broadcasts as a percentage of the overall network traffic. you would not be able to see that the number of broadcasts had increased. you are going to be disappointed. There is no absolute right answer to this question. etc. but should give you an idea of how a wide range of factors can affect your bandwidth patterns. honor system. • The number. you can detect anything from virus activity to spam email. These changes may in turn indicate a problem. excessive broadcasts. denial of service attacks. some of the factors that can influence the appearance of your traffic patterns are: • The capacity of your Internet connection • The number of users that have access to the network • The social policy (byte charging. Without an idea of what is "normal" for your network (a baseline). This is useful because it allows you to identify changes over time.sourceforge. let's look at the topic of baselines. While every environment is different. etc. If this graph shows a sudden increase in the amount of broadcast traffic. Establishing a baseline Since every environment is different. either sudden or gradual.net/. suppose that your network grinds to a halt.

25 shows traffic on an Aidworld firewall over a period of 24 hours. In Figure 6. you can predict the growth of your network and make changes before problems develop. MRTG. and Cacti are excellent tools you can use to keep a baseline. Figure 6. It is often very useful to experiment with such changes by trying different possible values. Knowing what the baseline looks like will show you whether your changes have improved matters. we can see the effect the implementation of delay pools has made on Internet utilization around the period of May. we would never know what the effect of the change over the long term was. Figure 6. say the top 100 sites accessed or the average utilization by your top twenty users.25: The traffic trend at Aidworld logged over a single day. If we did not keep a graph of the line utilization. or made them worse.24: By collecting data over a long period of time. RRDtool. but users were complaining about slow Internet access. Figure 6. to determine if habits have simply changed.24. You could then combine this baseline with others. As we will see later. When watching a total traffic graph after making changes. and was normally fin- . You might have removed frivolous usage from your line only to have it replaced by genuine legitimate traffic. There is nothing apparently wrong with this graph. don't assume that just because the graph does not change radically that your efforts were wasted. Figure 6.204 Chapter 6: Security & Monitoring Baseline graphs and figures are also useful when analyzing the effects of changes made to the network.26 shows that the upload bandwidth use (dark area) was higher during working hours on the last day than on previous days. A period of heavy upload usage started every morning at 03:00.

Figure 6.27: Four hours of jitter and packet loss.27 shows measurements of latency on the same connection as measured by a program called SmokePing. but that there were significant increases in latency at several times during the early morning. The color of the dots indicates the number of lost packets. which is probably heavy traffic of some kind. This indicates that additional monitoring should be performed during these early morning periods to establish the cause of the high latency.28) shows the same data over a period of 16 hours. up to 30 times the baseline value. This graph over a period of four hours does not help to identify whether there are any problems on the network. but on the last day it was still running at 16:30. Further investigation revealed a problem with the backup software. which ran at 03:00 every day. Figure 6. Figure 6. . which caused unexpected congestion for network users. The next graph (Figure 6.Chapter 6: Security & Monitoring 205 ished by 09:00. while the gray smoke indicates the distribution of latency (jitter).26: The same network logged over an entire week reveals a problem with backups. The position of the dots shows the average latency. This indicates that the values in the graph above are close to the normal level (baseline).

while the blue line indicates outbound traffic. Figure 6.206 Chapter 6: Security & Monitoring Figure 6. Depending on what sort of network environment you have.29 shows that Tuesday was significantly worse than Sunday or Monday for latency. For example. while monitoring cli- . Figure 6. This might indicate that something has changed on the network. the green area indicates inbound traffic.28: A higher spread of jitter is revealed in the 16 hour log. Inbound traffic is traffic that originates from another network (typically the Internet) and is addressed to a computer inside your network. monitoring of servers usually reveals larger amounts of outbound traffic as the servers respond to requests (such as sending mail or serving web pages). How do I interpret the traffic graph? In a basic network flow graph (such as that generated by the network monitor MRTG). the graph will help you understand how your network is actually being used. Outbound traffic is traffic that originates from your network.29: Zooming out to the week long view reveals a definite repetition of increased latency and packet loss in the early morning hours. and is addressed to a computer somewhere on the Internet. especially during the early morning period.

and is overutilized during these times. . It is up to you to establish a baseline to understand what normal network traffic patterns look like on your network. A router will normally show more incoming traffic than outgoing traffic as users download data from the Internet. There are no set metrics that indicate what outgoing traffic to incoming traffic should look like. An excess of outbound bandwidth that is not transmitted by your network servers may indicate a peer-to-peer client.31: Flat-topped graphs indicate that a line is using the maximum available bandwidth. Figure 6. or even a virus on one or more of your clients.31 shows traffic on an overloaded Internet connection. Detecting network overload Figure 6.Chapter 6: Security & Monitoring 207 ent machines might reveal higher amounts of inbound traffic to the machines as they receive data from the servers.30: The classic network flow graph. while the line represents outbound traffic. Figure 6. The repeating arcs of outbound traffic show when the nightly backups have run. The dark area represents inbound traffic. Traffic patterns will vary with what you are monitoring. unauthorized server.

or in the worst case. In this case it may indicate that you are not getting as much bandwidth from your service provider as you expect. The 95th percentile is a good value to use to show the bandwidth that is actually used at least 95% of the time. Its value shows the highest consumption of traffic for a given period.208 Chapter 6: Security & Monitoring The most apparent sign of overloading is the flat tops on outbound traffic during the middle of every day. Flat tops may indicate overloading. it is important to keep . Figure 6. may be incapable of responding at all. servers must have sufficient hardware capabilities to accommodate the work load. MRTG and Cacti will calculate the 95th Percentile for you. and processing power to accommodate the number of client requests. Monitoring RAM and CPU usage By definition. This is a sample graph of a 960 kbps connection. storage. providing access to services that are the whole point of having a network in the first place. the server will take longer to respond. Calculating the 95th percentile means that 95% of the time the usage is below a certain amount. and 5% of the time usage is above that amount. Measuring 95th percentile The 95th percentile is a widely used mathematical calculation to evaluate regular and sustained utilization of a network pipe. The 95th percentile came to 945 kbps after discarding the highest 5% of traffic. servers provide critical services that should always be available.32: The horizontal line shows the 95th percentile amount. Therefore. Since hardware resources are finite. This means they must have adequate RAM. Otherwise. even if they are well below the maximum theoretical capacity of the link. Servers receive and respond to client machine requests.

or other system resources cross a critical threshold. The simplest method on a Windows machine is to access the Task Manager using the Ctrl Alt + Del keys. Users will find a way to fill any size disk more quickly than you might think. There are several programs that can be used to monitor resources on a server.33: RRDtool can show arbitrary data. access times become slow. the mail server cannot receive mail. On a Linux or BSD box. If the disk or partition used for mail storage fills up. Nagios (see Page 200) can notify you when disk usage. To keep historical logs of such performance. all kinds of system problems may occur as the operating system runs out of swap space and temporary storage. Mail servers require adequate space. The messages can accumulate and fill the hard disk. This is often perceived by users as a network problem. . File servers need to be monitored. and then click on the Performance tab. even if they have large disks. MRTG or RRDtool (on Page 190) can also be used.Chapter 6: Security & Monitoring 209 track of how system resources are being used. If that disk is also used by the system. you can type top in a terminal window. or by simply monitoring usage and telling people when they are using too much. especially if quotas are not in use. such as memory and CPU usage. Figure 6. If a core server (such as a proxy server or email server) is overwhelmed by requests. Disk usage can be enforced through the use of quotas. expressed as an average over time. as some people may prefer to leave their email messages on the server for long periods of time. CPU utilization.

be sure to weigh the cost of parts vs. Some computers are difficult or impossible to upgrade. including shipping and taxes. Failing that. when determining the cost of upgrading. Slow speeds could also be as a result of insufficient RAM. Be sure to check the overall usage of CPU. and measurements show that a system resource is being heavily used. .210 Chapter 6: Security & Monitoring If a machine becomes unresponsive or slow. When the light is on constantly. You should always determine whether it is more cost effective to upgrade an individual component or purchase a whole new machine. It can usually be fixed by investigating which process is using the most RAM. it usually means that the machine is constantly swapping large amounts of data to and from the disk. RAM. and killing or reconfiguring that process. and it often costs more to replace individual components than to replace the entire system. Since the availability of parts and systems varies widely around the world. it may be time to upgrade the processor. This is known as thrashing. If processor usage constantly exceeds 60% of the total. A simple way to check whether a machine has insufficient RAM is to look at the hard disk light. this may be an indication that an upgrade is required. and disk space before deciding to upgrade a particular component. whole systems. and is extremely bad for performance. the system needs more RAM.

the precise amount of radiation that arrives at the surface of the Earth cannot be predicted with high precision. In this chapter. This chapter only discusses the use of solar energy for the direct production of electricity (photovoltaic solar energy). so it is common to talk about total irradiance per hour. Of course.7 Solar Power This chapter provides an introduction to the components of a standalone photovoltaic system. Therefore it is necessary to work with statistical data based on the "solar history" of a particular place. we will present the basic concepts of the generation and storage of photovoltaic solar energy. We will also provide a method for designing a functional solar system with limited access to information and resources. as tables or 211 . due to natural weather variations. Solar energy can also be used to heat fluids (thermal solar energy) which can then be used as a heat source or to turn a turbine to generate electricity. The word standalone refers to the fact that the system works without any connection to an established power grid. This data is gathered by a weather station over a long period and is available from a number of sources. Solar energy A photovoltaic system is based on the ability of certain materials to convert the radiant energy of the sun into electrical energy. Thermal solar energy systems are beyond the scope of this chapter. The instantaneous values are normally averaged over a period of time. day or month. The total amount of solar energy that lights a given area is known as irradiance (G) and it is measured in watts per square meter (W/m2).

it can be difficult to find detailed information about a specific area. Generally speaking. You can use the PSH value for your region to simplify your calculations. The peak sun hours are an easy way to represent the worst case average of irradiation per day. Low resolution PSH maps are available from a number of online sources.212 Chapter 7: Solar Power databases. The panels are responsible for collecting the energy of the sun and generating electricity. the regulator. To be effective. In most cases. and is the sum of the consumption of all . wind generators are able to charge batteries even at night. and you will need to work with approximate values. It is of course possible to use wind in conjunction with solar power to help cover times when there is extended cloud cover. or when there is insufficient wind. If we find that certain area has 4 PSH in the worst of the months. the average wind speed over the year should be at least 3 to 4 meter per second. A few organizations have produced maps that include average values of daily global irradiation for different regions. photovoltaic systems are more reliable than wind generators. A location far away from the coast usually lacks sufficient wind energy to support a wind powered system. consult a local solar energy vendor or weather station. as sunlight is more available than consistent wind in most places. it means that in that month we should not expect a daily irradiation bigger than 4000 W/m2 (day). the batteries. These values are known as peak sun hours or PSHs. and the load.com/solar-power-global-maps. The regulator ensures that panel and battery are working together in an optimal fashion. For more detailed information. and the wind generator should be 6 meters higher than other objects within a distance of 100 meters. The battery stores the electrical energy for later use. One unit of "peak sun" corresponds to a radiation of 1000 Watts per square meter. Photovoltaic system components A basic photovoltaic system consists of four main components: the solar panel. the cost of a good wind generator is not justified by the meager amount of power it will add to the overall system.html.solar4power. For most locations. What about wind power? It is possible to use a wind generator in place of solar panels when an autonomous system is being designed for installation on a hill or mountain. This chapter will therefore focus on the use of solar panels for generating electricity. On the other hand. as long as there is sufficient wind. The load refers to any device that requires electrical power. such as http://www.

Solar panel arrays can be made by connecting a set of panels in series and/or parallel in order to provide the necessary energy for a given load. This part of the system is sometimes referred to as a solar module or photovoltaic generator. Figure 7. These devices include proper wiring. If the range of operational voltage of your equipment does not fit the voltage supplied by your battery.Chapter 7: Solar Power 213 electrical equipment connected to the system. Amorphous silicon can be cheaper but is less efficient at converting solar . you will need to use a DC/DC converter. circuit breakers. This will vary according to climatological conditions. ground rods. Every electrical system should also incorporate various safety devices in the event that something goes wrong. If some of your equipment requires AC power. fuses. It is important to remember that solar panels and batteries use direct current (DC). the hour of the day. and the time of the year. The solar panel The solar panel is composed of solar cells that collect solar radiation and transform it into electrical energy. also known as an inverter. surge protectors. it will also be necessary to include some type of converter. and can be either monocrystalline or polycrystalline.1: A solar panel Several technologies are used in the manufacturing of solar cells. etc. If the equipment that you want to power uses a different DC voltage than the one supplied by the battery. The electrical current supplied by a solar panel varies proportionally to the solar radiation. you will need to use a DC/AC converter. lighting arrestors. The most common is crystalline silicon.

amorphous silicon is typically used for low power equipment. such as silicon ribbon and thin film photovoltaics. also called recombinant or VRLA (valve regulated lead acid) batteries.2: A 200 Ah lead-acid battery. New solar technologies. Figure 7. such as portable calculators. This stored energy can then be used during periods of low solar irradiation. These technologies promise higher efficiencies but are not yet widely available. Batteries store electricity in the form of chemical energy. The battery The battery stores the energy produced by the panels that is not immediately consumed by the load. This instantaneous power is needed to start some appliances. Aside from storing energy. . The negative terminal was broken due to weight on the terminals during transportation. sealed lead-acid batteries also serve two important functions: • They are able to provide an instantaneous power superior to what the array of panels can generate. such as the motor of a refrigerator or a pump. are currently under development. • They determine the operating voltage of your installation. With a reduced life expectancy and a 6 to 8% transformation efficiency.214 Chapter 7: Solar Power energy to electricity. The most common type of batteries used in solar applications are maintenance-free lead-acid batteries. The battery component is also sometimes called the accumulator.

timers. measurement of ampere-hour. These features include ammeters.3: A 30 Amp solar charge controller The regulator can include other features that add valuable information and security control to the equipment. A direct/alternating (DC/AC) converter. Figure 7. voltmeters. The SoC is estimated based on the actual voltage of the battery. By measuring the battery voltage and being programmed with the type of storage technology used by the battery. converts . other type of batteries (such as NiCd. the solar power charge regulator) assures that the battery is working in appropriate conditions. both of which are very detrimental to the life of the battery. The regulator The regulator (or more formally. etc. To ensure proper charging and discharging of the battery. The voltage provided might not match what is required by your load. none of these features are required for a working photovoltaic system. These types of batteries need a specialized charger/regulator and cannot directly replace lead-acid batteries. It avoids overcharging or overdischarging the battery. alarms. also known as inverter. While convenient. the regulator can know the precise points where the battery would be overcharged or excessively discharged. NiMh. or Li-ion) can be used. The converter The electricity provided by the panel array and battery is DC at a fixed voltage.Chapter 7: Solar Power 215 For a small power installation and where space constraints are important. the regulator maintains knowledge of the state of charge (SoC) of the battery.

For optimal operation. This comes at the price of losing some energy during the conversion. etc. If necessary. the system will support itself for years.4: An 800 Watt DC/AC converter (power inverter) The load The load is the equipment that consumes the power generated by your energy system. When all of the components are in balance and are properly maintained. Finally.216 Chapter 7: Solar Power the DC current from your batteries into AC. . the load consumes the stored energy to do work. The load may include wireless communications equipment. routers. The solar panels generate power when solar energy is available. DC/DC converters also lose some energy during the conversion. TV sets. lamps. you should design your solar-powered system to match the generated DC voltage to match the load. it is vital to be able to make a good estimate. VSAT modems. The regulator ensures the most efficient operation of the panels and prevents damage to the batteries. workstations. Converters and inverters adapt the stored energy to match the requirements of your load. In this type of system it is absolutely necessary to use efficient and low power equipment to avoid wasting energy. you can also use converters to obtain DC at voltage level other than what is supplied by the batteries. Figure 7. Although it is not possible to precisely calculate the exact total consumption of your equipment. Putting it all together The complete photovoltaic system incorporates all of these components. The battery bank stores collected energy for later use.

Chapter 7: Solar Power 217 + + Regulator - DC/DC + - DC load Solar panels + - Inverter AC load Battery bank Figure 7. and encapsulated between glass and plastic ma- . depending on the power demands of your application.5: A solar installation with DC and AC loads We will now examine each of the individual components of the photovoltaic system in greater detail. Figure 7. The most common modules are composed of 32 or 36 solar cells of crystalline silicon. wired in series. The solar panel An individual solar panel is made of many solar cells. These cells are all of equal size. The individual cells are properly encapsulated to provide isolation and protection from humidity and corrosion.6: The effect of water and corrosion in a solar panel There are different types of modules available on the market. The cells are electrically connected to provide a particular value of current and voltage.

but reduce the maximum voltage of the panel.5 m2. which represents the current that is provided based on the voltage generated for a certain solar radiation. shaded cells behave as a load that dissipates energy. The current (A) changes with the irradiance. Solar panels usually have two electrical contacts.) Bypass diodes will prevent hot-spots on shaded cells. shaded cells can see a significant increase in temperature (about 85 to 100ºC. and the voltage (V) changes with the temperature. In this situation.218 Chapter 7: Solar Power terial. They should only be used when shading is unavoidable. .7: Different IV Curves.1 and 0. The electrical performance of a solar module its represented by the IV characteristic curve. The surface area of the module is typically between 0. one positive and one negative. Bypass diodes protect the panel against a phenomenon known as “hot-spots”. using a polymer resin (EVA) as a thermal insulator. Some panels also include extra contacts to allow the installation of bypass diodes across individual cells. Rather than producing energy. A hot-spot occurs when some of the cells are in shadow while the rest of the panel is in full sun. Irradiance: 1kW / m2 Cell Temperature 25C 8 6 800W / m2 75C 50C 25C Current (A) 600W / m2 4 400W / m2 2 200W / m2 0 10 Voltage (V) 20 30 Figure 7. It is a much better solution to expose the entire panel to full sun whenever possible.

This is normally around 10-12%. as the voltage of operation is fixed by the load or the regulator. the more power a panel can provide. 5. depending on the type of cells (monocrystalline. EFFICIENCY (h): the ratio between the maximum electrical power that the panel can give to the load and the power of the solar radiation (PL) incident on the panel. ISC . and is directly proportional to the number of cells connected in series.8. Solar Panel Parameters The main parameters that characterize a photovoltaic panel are: 1.Chapter 7: Solar Power 219 The curve represents all the possible values of voltage-current. For a given solar cell area. FILL FACTOR (FF): the relation between the maximum power that the panel can actually provide and the product ISC . A good regulator will try to maximize the amount of energy that a panel provides by tracking the point that provides maximum power (V x I). VOC / PL . It is important not to forget that in normal conditions the panel will not work at peak conditions. The maximum power point of a panel is measured in Watts (W) or peak Watts (Wp). OPEN CIRCUIT VOLTAGE (VOC): the maximum voltage that the panel provides when the terminals are not connected to any load (an open circuit). VOC.7 and 0. Typical values of Vmax and Imax should be a bit smaller than the ISC and VOC 4. The closer FF is to 1. while the voltage reduces slightly with an increase of temperature. The maximum power corresponds to the knee of the IV curve. 2. polycrystalline. Considering the definitions of point of maximum power and the fill factor we see that: h = Pmax / PL = FF . where Pmax = Imax x Vmax. This gives you an idea of the quality of the panel because it is an indication of the type of IV characteristic curve. amorphous or thin film). 3. SHORT CIRCUIT CURRENT (ISC): the maximum current provided by the panel when the connectors are short circuited. MAXIMUM POWER POINT (Pmax): the point where the power supplied by the panel is at maximum. the current generated is directly proportional to solar irradiance (G). This value is normally 22 V for panels that are going to work in 12 V systems. The curves depend on two main factors: the temperature and the solar radiation received by the cells. Common values usually are between 0.

You should always be aware that the panel is not going to perform under perfect conditions as the load or regulation system is not always going to work at the point of maximum power of the panel. The panels are interconnected in such a way that the voltage generated is close to (but greater than) the level of voltage of the batteries. Connecting panels in parallel increases the current. Be aware that two panels can have the same Wp but very different behavior in different operating conditions. you should use panels of the same brand and characteristics because any difference in their operating conditions will have a big impact on the health and performance of your system. if possible. Using a solar panel array allows you to generate greater voltage and current than is possible with a single solar panel. Interconnection of panels A solar panel array is a collection of solar panels that are electrically interconnected and installed on some type of support structure. The panel parameters values change for other conditions of irradiance and temperature. ISC and VOC) match the values promised by the manufacturer. Panel parameters for system sizing To calculate the number of panels required to cover a given load. You should check the performance values at the panel temperatures that are likely to match your particular installation. When acquiring a panel. IPmax and VPmax are provided by the manufacturer and refer to standard conditions of measurement with irradiance G = 1000 W/m2. and that the current generated is sufficient to feed the equipment and to charge the batteries. for a temperature of cells of Tc = 25ºC. You should assume a loss of efficiency of 5% in your calculations to compensate for this. it is important to verify. that their parameters (at least. Connecting solar panels in series increases the generated voltage. Even panels that have identical . Manufacturers will sometimes include graphs or tables with values for conditions different from the standard. It is very important that all of the panels in your array are as identical as possible. you just need to know the current and voltage at the point of maximum power: IPmax and VPmax. In an array. VOC. The number of panels used should be increased until the amount of power generated slightly exceeds the demands of your load.220 Chapter 7: Solar Power The values of ISC. at sea-level.

This will give you a rough idea of the cost per Watt for different panels. Whenever possible. The voltage remains constant while the current duplicates. (Photo: Fantsuam Foundation. The actual operating characteristics of two panels from the same manufacturer can vary by as much as ±10%. If you are going to install solar panels in geographical areas where soiling (from dust. consider purchasing panels with a low affinity for soil retention.Chapter 7: Solar Power 221 performance ratings will usually display some variance in their characteristics due to manufacturing processes. Always check the mechanical construction of each panel. The solar . These panels are made of materials that increase the likelihood that the panel will be automatically cleaned by wind and rain. Figure 7. it is a good idea to test the real-world performance of individual panels to verify their operating characteristics before assembling them into an array. Verify that the glass is hardened and the aluminum frame is robust and well built. But there are a number of other considerations to keep in mind as well. or grit) will likely be a problem. sand.8: Interconnection of panels in parallel. Nigeria) How to choose a good panel One obvious metric to use when shopping for solar panels is to compare the ratio of the nominal peak power (Wp) to the price.

These cycles of charge and discharge occur whenever the energy produced by the panels does not match the energy required to support the load. thereby discharging it. any demand of electrical energy is supplied by the battery. A battery is formed by a set of elements or cells arranged in series. The battery experiences a cyclical process of charging and discharging. The batteries will also discharge when the irradiance is insufficient to cover the requirements of the load (due to the natural variation of climatological conditions. During the hours that there is sun.) . Look for the manufacturer's quality guarantee in terms of expected power output and mechanical construction. The battery serves two important purposes in a photovoltaic system: to provide electrical energy to the system when energy is not supplied by the array of solar panels.222 Chapter 7: Solar Power cells inside the panel can last for more than 20 years. be sure that the manufacturer provides not only the nominal peak power of the panel (Wp) but also the variation of the power with irradiation and temperature. The battery The battery “hosts” a certain reversible chemical reaction that stores electrical energy that can later be retrieved when needed. During the hours of absence of sun. and the reverse happens when the battery is discharged. dust. The energy that is not consumed immediately it is used to charge the battery. etc. When there is sufficient sun and the load is light. A potential difference of about 2 volts takes place between the electrodes. A 12 V battery therefore contains 6 cells in series. Leadacid batteries consist of two submerged lead electrodes in an electrolytic solution of water and sulfuric acid. depending on the presence or absence of sunlight. but they are very fragile and the panel must protect them from mechanical hazards. the array of panels produces electrical energy. as variations in the operating parameters can have a big impact on the quality of power generated and the useful lifetime of the panels. The most common batteries in photovoltaic solar applications have a nominal voltage of 12 or 24 volts. This is particularly important when panels are used in arrays. Finally. and to store excess energy generated by the panels whenever that energy exceeds the load. clouds. the batteries will charge. the batteries will discharge at night whenever any amount of power is required. depending on the instantaneous value of the charge state of the battery. Obviously. Electrical energy is transformed into chemical energy when the battery is being charged.

In the case of a telecommunications system. the oversizing the system (by adding far too many panels and batteries) is expensive and inefficient. good and cheap stationary batteries (importing batteries is not cheap). On the other hand. In any case. you will likely want to design your photovoltaic system with an autonomy of up to 5-7 days. One way to do this is to estimate the required number of days of autonomy. the system will be exhausted and will be unavailable for consumption. you will always have to find the proper balance between cost and reliability. Stationary batteries can use an electrolyte that is alkaline (such as NickelCadmium) or acidic (such as Lead-Acid). Traction batteries can be used in small applications where low cost is the most important consideration. Stationary batteries based on Nickel-Cadmium are recommended for their high reliability and resistance whenever possible. The most suitable type for photovoltaic applications is the stationary battery. or when other batteries are not available.Chapter 7: Solar Power 223 If the battery does not store enough energy to meet the demand during periods without sun. Using car batteries Automobile batteries are not well suited for photovoltaic applications as they are designed to provide a substantial current for just few seconds (when starting then engine) rather than sustaining a low current for long period of time. In areas with low irradiance. and are intended for use in a variety of different applications. "Stationary" batteries can accommodate deep discharge cycles. On the other hand. designed to have a fixed location and for scenarios where the power consumption is more or less irregular. When designing a stand-alone system we need to reach a compromise between the cost of components and the availability of power from the system. but they are not designed to produce high currents in brief periods of time. you will be forced to use batteries targeted to the automobile market. this value may need to be increased even more. Unfortunately. . If the equipment is going to serve as repeater and is part of the backbone of your network. if the solar system is responsible for a providing energy to client equipment you can probably reduce number of days of autonomy to two or three. the number of days of autonomy depends on its critical function within your network design. This design characteristic of car batteries (also called traction batteries) results in an shortened effective life when used in photovoltaic systems. Types of batteries Many different battery technologies exist. they tend to be much more expensive and difficult to obtain than sealed lead-acid batteries. In many cases when it is difficult to find local.

in a process is known as gasification. These batteries should never be deeply discharged. because doing so will greatly reduce their ability to hold a charge.28. the acid tends to concentrate itself at the bottom of the battery thereby reducing the effective capacity. the presence of gas avoids the stratification of the acid. oxidation on the positive electrode.4 Volts for each element of the battery every few days. Again. at 25ºC. you must use distilled water. This means that you can only use a maximum of 30% of a lead-acid battery's nominal capacity before it must be recharged. One solution is to allow a slight overcharge condition every so often. and in extreme cases. at a cost of reducing the overall capacity of the battery. On the other hand. the electrolyte begins to break down. The regulator should ensure a periodical and controlled overcharges. After several continuous cycles of charge and discharge. a danger of explosion. They are cheaper than stationary batteries and can serve in a photovoltaic installation.2 can help reduce the anode's corrosion. By using a densimeter or hydrometer. This results in a loss of water. you can measure the density of the battery's electrolyte. States of charge There are two special state of charge that can take place during the cyclic charge and discharge of the battery. If you adjust the density of battery electrolyte. This produces bubbles of oxygen and hydrogen.35 to 2.224 Chapter 7: Solar Power Traction batteries are designed for vehicles and electric wheelbarrows. as tap water or well water will permanently damage the battery. A truck battery should not discharged by more than 70% of its total capacity. it is necessary to find a compromise between the advantages (avoiding electrolyte stratification) and the disadvantages (losing water and production of hydrogen). One typical method is to allow a voltage of 2. Adding distilled water and lowering the density to 1. Overcharge Overcharge takes place when the battery arrives at the limit of its capacity. If energy is applied to a battery beyond its point of maximum charge. A typical battery has specific gravity of 1. You can extend the life of a lead-acid battery by using distilled water. . although they require very frequent maintenance. The process of gasification agitates the electrolyte and avoids stratification. They should both be avoided in order to preserve the useful life of the battery.

the regulator disconnects the load from the battery. It is equal to the product of the nominal capacity and the maximum DoD. expressed as a percentage. Traction batteries should only be discharged by as little as 30%. • Useful Capacity. • Nominal Capacity. three effects take place: the formation of crystallized sulfate on the battery plates. For example. the most common value being 12 V. . Discharging beyond that limit will result in deterioration of the battery. When the effective battery supply is exhausted. When the voltage of the battery reaches the minimum limit of 1. CUBat: It is the real (as in usable) capacity of a battery. Discharging a battery over a long period will yield more energy compared to discharging the same battery over a short period. • Maximum Depth of Discharge. DoDmax: The depth of discharge is the amount of energy extracted from a battery in a single discharge cycle. The life expectancy of a battery depends on how deeply it is discharged in each cycle. It is expressed in Ampere-hours (Ah) or Watt-hours (Wh). and plate buckling. CNBat: the maximum amount of energy that can be extracted from a fully charged battery. a stationary battery of nominal capacity (C100) of 120 Ah and depth of discharge of 70% has a useful capacity of (120 x 0. This is particularly negative as it generates big crystals that do not take part in any chemical reaction and can make your battery unusable. this time should be longer than 100 hours (C100). The amount of energy that can be obtained from a battery depends on the time in which the extraction process takes place. The process of forming stable sulfate crystals is called hard sulfation. For photovoltaic applications.7) 84 Ah. there is also a lower limit to a battery's state of charge. Battery Parameters The main parameters that characterize a battery are: • Nominal Voltage. If the discharge of the battery is very deep and the battery remains discharged for a long time. As a general rule you should avoid discharging a deep cycle battery beyond 50%.Chapter 7: Solar Power 225 Overdischarge In the same way that there is a upper limit. The capacity of a battery is therefore specified at different discharging times. the loosening of the active material on the battery plate.85 Volts per cell at 25°C. the regulator prevents any more energy from being extracted from the battery. VNBat. The manufacturer should provide graphs that relate the number of charge-discharge cycles to the life of the battery.

Using the same table.58 11.6 V depending on the state of charge.6 to 11. etc. the temperature.226 Chapter 7: Solar Power Measuring the state of charge of the battery A sealed lead-acid battery of 12 V provides different voltages depending on its state of charge.06 11.6 V when "empty".42 12.93 1. When the battery is fully charged in an open circuit.96 1.9 V.7 12.6 V when "full" and 11.03 2.9 11.75 11. .98 1. State of Charge 12V Battery Voltage 12.01 1. we find that we should program the regulator to prevent the battery from discharging below 12. the output voltage is about 12.08 2.89 1.2 12.6 V when loads are attached. we can determine that the useful capacity of a truck 170 Ah truck battery is 34 Ah (20%) to 51 Ah (30%). As the battery is providing constant current during operation.3 V.12 2. These values are only a rough approximation since they depend on the life and quality of the battery.32 12.8 V. If we make the broad assumption that a fully loaded battery has a voltage of 12. the battery voltage reduces linearly from 12.05 2.5 12. we can estimate that a battery has discharged 70% when it reaches a voltage of 11. A sealed lead-acid batteries provides 95% of its energy within this voltage range.31 10.5 Volts per Cell 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2.07 2. and considering that a truck battery should not be discharged more than 20% to 30%. The output voltage lowers quickly to 12.75 According to this table.

The maximum current of the fuse should be 20% bigger than the maximum current expected. to protect the battery from short-circuit in case of regulator failure. so two different kinds of protection needs to be considered. and quick blow. a short circuit can lead to a very high current which can easily reach several hundred amperes. The regulator is connected to the battery and the loads. and possibly cause electric shock to a human body If a fuse breaks. Even if the batteries carry a low voltage. damage the equipment and batteries. Quick blow fuses will immediately blow if the current flowing through them is higher than their rating. then replace the fuse with another one which has the same characteristics. currents reach levels of 45 A during charging Every fuse is rated with a maximum current and a maximum usable voltage. Slow blow fuses should be used with inductive or capacitive loads where a high current can occur at power up. Slow blow fuses will allow a higher current than their rating to pass for a short time. never replace a fuse with a wire or a higher rated fuse. A second fuse is needed to protect the regulator from excessive load current. Large currents can cause fire. . Figure 7. One fuse should be placed between the battery and the regulator. This second fuse is normally integrated into the regulator itself.Chapter 7: Solar Power 227 Battery and regulator protection Thermomagnetic circuit breakers or one time fuses must be used to protect the batteries and the installation from short circuit and malfunctions.9: A battery bank of 3600 Ah. First determine the cause of the problem. There are two types of fuses: slow blow.

It's also desirable to place the batteries on a small support to allow air to flow under them. This problem can be compensated partially in car batteries by using a low density of dissolution (a specific gravity of 1. the chemical reaction that takes place in the battery accelerates. As laboratory tests are . In areas of low temperatures. the greater the risk of freezing. A good battery should always come with its technical specifications.228 Chapter 7: Solar Power Temperature effects The ambient temperature has several important effects on the characteristics of a battery: • The nominal capacity of a battery (that the manufacturer usually gives for 25°C) increases with temperature at the rate of about 1%/°C. you run the the risk of freezing the electrolyte. and battery terminals should be free of corrosion. and requirements for chargers. you should avoid deeply discharging the batteries (that is. It is preferable to use a regulator which adjusts the low voltage disconnect and reconnect parameters according to temperature. liquid spillage or any sign of damage. The batteries must be free of cracks. DoDmax is effectively reduced. The temperature sensor of the regulator should be fixed to the battery using tape or some other simple method. thus increase the cooling. bulky and expensive to import.) • The temperature also changes the relation between voltage and charge. High capacity batteries are heavy. But if the temperature is too high. But if the temperature is too low.25 when the battery is totally charged). operating temperature. How to choose a good battery Choosing a good battery can be very challenging in developing regions. which is also related to the state of charge of the battery. the useful life of the battery increases. A 200 Ah battery weights around 50 kg (120 pounds) and it can not be transported as hand luggage. • In hot areas it is important to keep the batteries as cool as possible. The batteries must be stored in a shaded area and never get direct sunlight. which can cause the same type of oxidation that takes places during overcharging. cut-off voltage points. If you want long-life (as in > 5 years) and maintenance free batteries be ready to pay the price. The lower the density. • As the temperature is reduced. The freezing temperature depends on the density of the solution. C100). This will obviously reduce the life expectancy of battery. including the capacity at different discharge rates (C20.

Even deep cycle batteries will have an increased battery life if the the number of deep discharge (>30%) cycles is reduced. The connection and disconnection is done by means of switches which can be of two types: electromechanical (relays) or solid state (bipolar transistor. The regulator sits between the array of panels. Remember that the voltage of a battery. charge-discharge controller or charge-discharge and load controller. A typical price (not including transport and import tax) is $3-4 USD per Ah for 12 V lead-acid batteries. the regulator prevents overcharging or overdischarging. The power charge regulator The power charge regulator is also known as charge controller. If you use only 1/3 of the capacity the battery. By monitoring the voltage of the battery. It can be cheaper to buy a battery with 3 times the capacity than to change the battery every year. the batteries. expect lots of low quality batteries (including fakes) in the local markets. varies according to its state of charge. MOSFET). Life expectancy versus number of cycles Batteries are the only component of a solar system that should be amortized over a short period and regularly replaced. you will typically need to change it after less than one year.Chapter 7: Solar Power 229 necessary to obtain complete data about real capacity and aging. If you completely discharge the battery every day. although always close to 2 V per cell. The low voltage disconnect (LVD) prevents the battery from overdischarging by disconnecting or shedding the load. In order to protect the battery from gasification. and your equipment or loads. Regulators should never be connected in parallel. Regulators used in solar applications should be connected in series: they disconnect the array of panels from the battery to avoid overcharging. voltage regulator. To prevent continuous connections and disconnections the regulator will not connect back the loads until the battery reaches a low reconnect voltage (LRV). You can increase the useful lifetime of a battery by reducing the depth of discharge per cycle. and they disconnect the battery from the load to avoid overdischarging. the switch opens the charging circuit when the voltage in the battery reaches its high voltage disconnect (HVD) or cut-off set point. . it can last more than 3 years.

. or 48 V.5 The most modern regulators are also able to automatically disconnect the panels during the night to avoid discharging of the battery. Regulator Parameters When selecting a regulator for your system. and they may use a mechanism known as pulse width modulation (PWM) to prevent excessive gassing. As the peak power operating point of the array of panels will vary with temperature and solar illumination.3 14. you should at least know the operating voltage and the maximum current that the regulator can handle. Some regulators include special alarms to indicate that the panels or loads have been disconnected. LRV and HVD.6 15. LVD or HVD has been reached.5 12. The maximum current must be 20% bigger than the current provided by the array of panels connected to the regulator. The voltage that indicates the state of charge of the battery vary with temperature. etc. For that reason some regulators are able to measure the battery temperature and correct the different cut-off and reconnection values. The most common instruments measure the voltage of the panels and batteries. new regulators are capable of constantly tracking the maximum point of power of the solar array. • Instrumentation and gauges. • Support for temperature compensation. Other features and data of interest include: • Specific values for LVD.6 14. 24. They can also periodically overcharge the battery to improve their life. the state of charge (SoC) or Depth of Discharge (DoD). This feature is known as maximum power point tracking (MPPT).230 Chapter 7: Solar Power Typical values for a 12 V lead-acid battery are: Voltage Point LVD LRV Constant Voltage Regulated Equalization HVD Voltage 11. The operating Voltage will be 12.

When it comes to efficiency. Inverters have two power ratings: one for continuous power. The efficiency of a linear regulator decreases as the difference between the input voltage and the output voltage increases. There are two conversion methods which can be used to adapt the voltage from the batteries: linear conversion and switching conversion. Motors will work. the linear regulator will have an efficiency of only 50%. or the inverse (negative) of the input voltage. The inverter should also be able to safely interrupt . A standard switching regulator has an efficiency of at least 80%. DC power supplies tend to warm up more. The resulting voltage can be greater. In addition. Most models available on the market produce what is known as "modified sine wave". some laser printers will not work with a modified sine wave inverter. DC/AC Converter or Inverter Inverters are used when your equipment requires AC power. Be aware that not all the equipment will accept a modified sine wave as voltage input. This method is very simple but is obviously inefficient. Most commonly. as their voltage output is not a pure sinusoid. Linear conversion lowers the voltage from the batteries by converting excess energy to heat. as when starting a motor. some important features of inverters include: • Reliability in the presence of surges. if we want to convert from 12 V to 6 V.Chapter 7: Solar Power 231 Converters The regulator provides DC power at a specific voltage. and audio amplifiers can emit a buzzing sound. Switching conversion generally uses a magnetic component to temporarily store the energy and transform it to another voltage. They are capable of providing the peak power for a very short amount of time. less than. Inverters chop and invert the DC current to generate a square wave that is later filtered to approximate a sine wave and eliminate undesired harmonics. For example. Very few inverters actually supply a pure sine wave as output. modified sine wave inverters perform better than pure sinusoidal inverters. Converters and inverters are used to adjust the voltage to match the requirements of your load. DC/DC Converters DC/DC converters transform a continuous voltage to another continuous voltage of a different value. Aside from the type of waveform. and a higher rating for peak power. but they may consume more power than if they are fed with a pure sine wave.

toasters. Most communications equipment can accept a wide range of input voltage. • It is possible to use photovoltaic power for appliances that require low and constant consumption (as in a typical case. LED lamps are also a good choice as they are very efficient and are fed with DC. the TV). solar) depending on what is available.) Whenever possible. Inverters are most efficient when providing 50% to 90% of their continuous power rating. the expense of the photovoltaic system also increases. Equipment or load It should be obvious that as power requirements increase. • Battery charging. • Automatic fall-over. they have much better energy efficiency than incandescent light bulbs. This type of inverter is known as a charger/ inverter. generator. Although these lamps are more expensive. it is best to avoid the use of DC/ AC converters and feed them directly from a DC source. When designing the system you must first make a realistic estimate of the maximum consumption. or if the requested power is too high.232 Chapter 7: Solar Power itself (with a circuit breaker or fuse) in the event of a short circuit. Home Appliances The use of photovoltaic solar energy is not recommended for heat-exchange applications (electrical heating. refrigerators. generator. etc). the established maximum consumption must be respected in order to avoid frequent power failures. Once the installation is in place. It is therefore critical to match the size of the system as closely as possible to the expected load. etc. You should select an inverter that most closely matches your load requirements. Here are some points to keep in mind when choosing appropriate equipment for use with a solar system: • The photovoltaic solar energy is suitable for illumination. Smaller televisions . energy should be used sparingly using low power appliances. Many inverters also incorporate the inverse function: the possibility of charging batteries in the presence of an alternative source of current (grid. the use of halogen light bulbs or fluorescent lamps is mandatory. The manufacturer usually provides the performance of the inverter at 70% of its nominal power. • Conversion efficiency. In this case. Some inverters can switch automatically between different sources of power (grid. When using telecommunication equipment.

reducing transmit power to the minimum value that is necessary to provide a stable . Here is a table that gives you a general idea of the power consumption that you can expect from different appliances. • If you must use a refrigerators. A well known saying of radio amateurs applies here. Further measures to reduce power consumption include throttling the CPU speed. • Photovoltaic solar energy is not recommended for any application that transforms energy into heat (thermal energy). A Wi-Fi card with good receiver sensitivity and a fresnel zone that is at least 60% clear will work better than an amplifier. it should consume as little power as possible. • Conventional automatic washing machines will work. The estimation of total consumption is a fundamental step in sizing your solar system. but you should avoid the use of any washing programs that include centrifuged water heating.Chapter 7: Solar Power 233 use less power than larger televisions. although their consumption can be quite high (around 1000 Wh/day). too: The best amplifier is a good antenna. For example . Also consider that a black-and-white TV consumes about half the power of a color TV. Use solar heating or butane as alternative. and save power consumption as well. Equipment Portable computer Low power lamp WRAP router (one radio) VSAT modem PC (without LCD) PC (with LCD) Network Switch (16 port) Consumption (Watts) 30-50 6-10 4-10 15-30 20-30 200-300 6-8 Wireless telecommunications equipment Saving power by choosing the right gear saves a lot of money and trouble. a long distance link doesn't necessarily need a strong amplifier that draws a lot of power. There are specialized refrigerators that work in DC.

If they are present then the device has a switched mode input. One of the boards with lowest power consumptions is the Soekris platform that uses an AMD ElanSC520 processor. WARNING: Operating your access point with a power supply other than the one provided by your manufacturer will certainly void any warranty. Open your access point and look near the DC input for two relatively big capacitors and an inductor (a ferrite toroid with copper wire wrapped around it). A router or access point that accepts 8-20 Volts DC is perfect. but it has an Ethernet switch onboard. Equipment based on traditional Intel x86 CPUs are power hungry in comparison with RISC-based architectures as ARM or MIPS. Another alternative to AMD (ElanSC or Geode SC1100) is the use of equipment with MIPS processors. opening your device will void any existing warranty. While the following technique will typically work as described. remember that should you attempt it. increasing the length of beacon intervals. MIPS processors have a better performance than an AMD Geode at the price of consuming between 20-30% of more energy. connecting an unregulated power supply with 24 Volts to a device with 25 Volt-capacitors is not a good idea. The 4G Systems Accesscube draws . Usually the rating of these capacitors is 16 or 25 volts. Most cheap access points have a switched mode voltage regulator inside and will work through such a voltage range without modification or becoming hot (even if the device was shipped with a 5 or 12 Volt power supply). The popular Linksys WRT54G runs at any voltage between 5 and 20 volts DC and draws about 6 Watts. Preferably. you do so at your own risk. Linksys also offers a Wi-Fi access point called WAP54G that draws only 3 Watts and can run OpenWRT and Freifunk firmware. Do not try to operate an access point at higher voltage if it doesn't have a switched mode regulator. malfunction. Transforming the voltage provided by the battery to AC or using a voltage at the input of the access point different from the voltage of the battery will cause unnecessary energy loss. Having a switch is of course nice and handy . Most autonomous solar systems work at 12 or 24 volts. operating at the 12 Volts that most lead acid batteries provide. So. Of course.but it draws extra power. It will get hot. or burn. and switching the system off during times it is not needed.234 Chapter 7: Solar Power link. a wireless device that runs on DC voltage should be used. Be aware that an unregulated power supply has a ripple and may feed a much higher voltage into your access point than the typical voltage printed on it may suggest. and may cause damage to your equipment. and the maximum input voltage should be somewhat below the voltage printed on the capacitors.

indicating pending traffic.8 2. a wireless board of low consumption consumes 2 to 3 W.11b is sufficient. The main mechanism for energy saving is to allow stations to periodically put their wireless cards to "sleep" by means of a timing circuit. its benefit is not as good as you might hope.04 2.Chapter 7: Solar Power 235 about 6 Watts when equipped with a single WiFi interface. High power cards (such as the 400 mW Ubiquity) consume around 6 W. As a general rule. radios. It is . mini-PCI cards with the Orinoco chipset perform very well while drawing a minimum amount of power.53 3 6 The amount of power required by wireless equipment depends not only on the architecture but on the number of network interfaces. type of memory/storage and traffic. When the wireless card wakes up it verifies if a beacon exists. Power saving mode may be incompatible between implementations from different manufacturers. which can cause unstable wireless connections.11 incorporates a power saving mode (PS) mechanism. and a 200 mW radio card consumes as much as 3 W. A repeating station with two radios can range between 8 and 10 W. Although the standard IEEE 802. as the access point always needs to remain awake to send beacons and store traffic for the clients.3 1.1E-1 (no radio) Mikrotik Routerboard 532 (no radio) Inhand ELF3 (no radio) Senao 250mW radio Ubiquiti 400mW radio Consumption (Watts) 6 3 15 1. The energy saving therefore only takes place in the client side. Equipment Linksys WRT54G (BCM2050 radio) Linksys WAP54G (BCM2050 radio) Orinoco WavePoint II ROR (30mW radio) Soekris net4511 (no radio) PC Engines WRAP. If 802.

Wire size is normally given in American Wire Gauge (AWG). Corroded connection have an increased resistance. The azimuth is the angle that measures the deviation with respect to the . • For low currents (<10 A) consider the use of Faston or Anderson powerpole connectors. resulting in loss. For bigger currents. During your calculations you will need to convert between AWG and mm2 to estimate cable resistance. the azimuth a and the inclination or elevation ß. The solar module will capture more energy if it is “facing” the sun.11 mm and can handle up to 55 A. as the difficulties created will likely outweigh the meager amount of saved power. including an estimate of resistance and current carrying capacity. Selecting the voltage Most low power stand-alone systems use 12 V battery power as that is the most common operational voltage in sealed lead-acid batteries. use metallic ring lugs. Keep in mind that the current carrying capacity can also vary depending on the type of insulation and application. perpendicular to the straight line between the position of the installation and the sun. Some good practices that you should consider include: • Use a screw to fasten the cable to the battery terminal. Loose connections will waste power. consult the manufacturer for more information. The orientation of the panels is determined by two angles. • Spread Vaseline or mineral jelly on the battery terminals. Wiring An important component of the installation is the wiring. When designing a wireless communication system you need to take into consideration the most efficient voltage of operation of your equipment. the sun's position is constantly changing relative to the earth. While the input voltage can accept a wide range of values. an AWG #6 cable has a diameter of 4. For example. so we need to find an optimal position for our panels. Orientation of the panels Most of the energy coming from the sun arrives in straight line. A conversion chart. Of course. When in doubt.236 Chapter 7: Solar Power nearly always best to leave power saving mode disabled on all equipment. as proper wiring will ensure efficient energy transfer. is available in Appendix D. you need to ensure that the overall power consumption of the system is minimal.

etc. You should use the absolute value of the latitude of the place (angle F) increased by 10° (ß = | F | + 10 °). It is acceptable to turn the panels ±20º towards the east or the west if needed (a = ±20º). The inclination is the angle formed by the surface of the module and the horizontal plane. It is very important that no part of the panels are ever under shade!. In areas where snow and ice . Ideally. • For installations with less consumptions during winter. The value of ß should maximize the ratio between the offer and the demand of energy.10°). it is preferable to optimize the installation to capture the maximum radiation during "the winter" months. The maximum height that the sun reaches every day will vary. The inclination of the panel should never be less than 15° to avoid the accumulation of dust and/or humidity on the panel. the parameter that is key in our calculations is the inclination of the panel. with the maximum on the day of the summer solstice and the minimum on the winter solstice. In most telecommunications scenarios the energy demands of the system are constant throughout the year. Providing for sufficient power during the "worst month" will work well for the rest of the year.Chapter 7: Solar Power 237 south in the northern hemisphere. walls. the panels should track this variation. In installations with telecommunications equipment it is normal to install the panels at a fixed inclination. Inclination Once you have fixed the azimuth. • For installations that are only used during summer. which we will express as the angle beta (ß). the value of the latitude of the place can be used as the solar panel inclination. Azimuth You should have the module turned towards the terrestrial equator (facing south in the northern hemisphere.) to be sure that they will not cast a shadow on the panels at any time of the day or year. but this is usually not possible for cost reasons. This way the system is optimized for the months of spring and autumn (ß = | F |). other panels. and with respect to the north in the southern hemisphere. • For installations with consistent (or nearly consistent) consumption throughout the year. and north in the southern) so that during the day the panel catches the greatest possible amount of radiation (a = 0). Study the elements that surround the panel array (trees. buildings. you should use the absolute value of the latitude of the place (angle F) decreased by 10° (ß = | F | .

it is never possible to be absolutely sure that a standalone system is going to be able to provide the necessary energy at any particular moment. money) is wasted. at which point no additional power will be available. If there is a considerable increase in consumption during the summer. at a minimum: • The number and type of solar panels required to capture enough solar energy to support your load. For example.e. Of course. solar radiation) can be difficult to predict. Never forget that the ability of the photovoltaic energy to produce and store electrical energy is limited. This would require special support structures and a regular schedule for changing the position of the panels. and based on them to calculate a system that works for the maximum amount of time so it is as reliable as possible. because unless the system components are balanced. can be a cause of failure (or even fire) and render the installation unusable. or one single cable that is too small. and will determine your number of days of autonomy.238 Chapter 7: Solar Power occur. System sizing calculations are important. you might consider arranging for two fixed inclinations. The battery will need to store enough energy to provide power at night and through days with little sun. In fact. it is very important to protect the panels and to incline them an angle of 65° or greater. if more . one position for the months of summer and another for the months of winter. The design method that we propose consists of considering the energy requirements. How to size your photovoltaic system When choosing equipment to meet your power needs. the batteries should have enough capacity to store the additional energy produced. If the bank of batteries is too small and the load is not using the energy as it is generated. if we install more solar panels to produce more energy.) needed to support the amount of power generated and stored. Solar systems are designed for a certain consumption. • The minimum capacity of the battery. and if the user exceeds the planned limits the provision of energy will fail. energy (and ultimately. Accidentally leaving on a light bulb during the day can easily drain your reserves before nighttime. wiring. • The characteristics of all other components (the regulator. A regulator of a smaller amperage than needed. The availability of "fuel" for photovoltaic systems (i. you will need to determine the following. etc. then energy must be thrown away.

low cost is likely going to be a the most important factor.Chapter 7: Solar Power 239 panels and batteries are installed. Finding a balance between cost and reliability is not a easy task. a hospital. Will it be used to illuminate houses. it is necessary to know the climatology of the place. This increase of reliability will also have an increase in cost. One commonly used value for critical telecommunications equipment is N = 5. as well as the economic and social relevance of the installation. more energy will be able to be collected and stored. Using this method. It is the worst month of the year. when all consumption is made solely at the expense of the energy stored in the battery. and many of them cannot be evaluated easily. and can be thought of as the number of consecutive cloudy days when the panels do not collect any significant amount of energy. The method we will use for sizing the system is known as the method of the worst month. we have included several tables that will facilitate the collection of required data for sizing the system. a factory. Your experience will play an important role in this part of the system sizing. whereas for low cost client equipment it is possible to reduce the autonomy to N = 3. and at what price. It is not the same to change a discharged battery from an installation in the middle of a city versus one at the top a telecommunication tower that is several hours or days of walking distance. We simply calculate the dimensions of the standalone system so it will work in the month in which the demand for energy is greatest with respect to the available solar energy. In Appendix E. or for some other application? Remember that as N increases. you should be able to determine what it is expected from your design choices. Fixing the value of N it is not an easy task as there are many factors involved. In some photovoltaic installations (such as the provision of energy for telecommunications equipment on a network backbone) the reliability factor is more important that the cost. When choosing N. so does the investment in equipment and maintenance. for a radio link. It is also important to evaluate all possible logistical costs of equipment replacement.) This is known as the maximum number of days of autonomy (N). but whatever your situation. In a client installation. reliability is taken into consideration by fixing the maximum number of days that the system can work without receiving solar radiation (that is. . The rest of this chapter will explain in detail what information you need to collect or estimate and how to use the method of the "worst month". as this month with have the largest ratio of demanded energy to available energy.

So when we refer to 5 peak sun hours. Remember to use a positive sign in the northern hemisphere and negative in the south.240 Chapter 7: Solar Power Data to collect • Latitude of the installation. . Do not confuse "sun hours" with the number of "peak sun hours". It is advisable to make your our own measurements to check for any deviation from the nominal values. in kWh/m2 per day). but refers to the amount of daily irradiation. You should check for this information from a weather station close to your implementation site. If you have the data in Joules (J). • Solar radiation data. this implies a daily solar radiation of 5000 W/m2. These are the minimum values that you need to gather before starting your system sizing: Panels You need to know the voltage VPmax and the current IPmax at the point of maximum power in standard conditions. you can apply the following conversion: 1 J = 2. The monthly value is the sum of the values of global irradiation for every day of the month. It is a good idea to ask companies that install photovoltaic systems in the region. one for every month. as their experience can be of great value. The number of peak sun hours has nothing to do with the number of hours without clouds.78 10-7 kWh The irradiation data Gdm(0) of many places of the world is gathered in tables and databases. For the method of the "worst month" it is enough to know just twelve values. A peak sun hour is a normalized value of solar radiation of 1000 W/m2 at 25 C. Unfortunately. divided by the number of days of the month. but do not be surprised if you cannot find the data in electronic format. The twelve numbers are the monthly average values of daily global irradiation on horizontal plane (Gdm(0). A day of 5 hours of sun without clouds does not necessary have those hours when the sun is at its zenith. deviation from promised values can be large and should be expected. Electrical characteristics of system components The electrical characteristics of the components of your system should be provided by the manufacturer.

etc.Chapter 7: Solar Power 241 Batteries Nominal capacity (for 100 hours discharge) CNBat . operational voltage VNBat . it is necessary to decide on two more pieces of information before being able to size a photovoltaic system. N. and consult with an experienced designer to reach a final decision. it is also very important to consider the average time each load will be used. In order to know the total energy that our installation is going to consume. AGM. weekly. whether sealed lead-acid. Equipment or load It is necessary to know the nominal voltage VNC and the nominal power of operation PC for every piece of equipment powered by the system. It is impossible to give a concrete value of N that is valid for every installation. These two decisions are the required number of days of autonomy and the operational voltage of the system. you need to know the nominal voltage VNConv. monthly or annually? Consider any changes in the usage that might impact the amount of energy needed (seasonal usage. Is it constant? Or will it be used daily. Take these values as a rough approximation. DC/AC Converter/Inverter If you are going to use a converter. instantaneous power PIConv and performance at 70% of maximum load H70. and either the maximum depth of discharge DoDmax or useful capacity CUBat . number of days of autonomy You need to decide on a value for N that will balance meteorological conditions with the type of installation and overall costs.) Other variables Aside from the electrical characteristics of the components and load. gel. modified traction etc. training or school periods. You also need to know the type of battery that you plan to use. The type of battery is important when deciding the cut-off points in the regulator. but the next table gives some recommended values. . Regulator You need to know the nominal voltage VNReg. and the maximum current that can operate ImaxReg.

Record the power consumption characteristics of the equipment chosen as well as estimated usage. reducing the statistical data to 12 values. and the orientation and the optimal inclination of the solar panels. 2. This estimation is a good compromise between precision and simplicity. . try to fix the nominal voltage to 12 or 24 V. • If you need to power several types of equipment that work at different nominal voltages. one for each month of the year. nominal voltage of the installation The components of your system need to be chosen to operate at a nominal voltage VN. The selection of VN is not arbitrary. You should consider the expected fluctuations of usage due to the variations between winter and summer. we calculate the solar energy available. etc. The result will be 12 values of energy demand. • If the equipment allows it. school / vacation periods. the voltage will be 48 V. The estimation of solar energy available is done in monthly intervals. Then calculate the electrical energy required on a monthly basis. Calculate the available solar energy (the offer). Based on statistical data of solar radiation. and depends on the availability of equipment.242 Chapter 7: Solar Power Available Sunlight Very cloudy Variable Sunny Domestic Installation 5 4 3 Critical Installation 10 8 6 VN. and if the total power of consumption surpasses 3 kW. This voltage is usually 12 or 24 Volts for small systems. Many wireless communications boards accept a wide range of input voltage and can be used without a converter. the rainy period / dry season. Estimate the required electrical energy (the demand). calculate the voltage that minimizes the overall power consumption including the losses for power conversion in DC/DC and DC/ AC converters. Procedure of calculation There are three main steps that need to be followed to calculate the proper size of a system: 1.

and due to the fact that the panels do not always work at the point of maximum power. and the total energy that you require ETOTAL(WORST MONTH) you can proceed to the final calculations.co.pvsyst. Calculate the ideal system size (the result). and the system sizing is based on the data of that worth month. in a day with a irradiation of Gdm for month "m". • The necessary energy storage capacity to cover the minimum number of days of autonomy. The calculations of Gdm(ß) for a certain place can be made based on Gdm(0) using computer software such as PVSYST (http://www. the total voltage is equal to the sum of the individual voltages of each module. • The length and the necessary sections of cables for the electrical connections. Required current in the worst month For each month you need to calculate the value of Im. • The required electrical characteristics of the regulator. ETOTAL is the sum of all DC and AC loads.Chapter 7: Solar Power 243 3. Number of panels By combining solar panels in series and parallel. which will determine the minimum number of panels. which is the maximum daily current that an array of panels operating at nominal voltage of VN needs to provide. When panels are connected in series.com/) or PVSOL (http://www. To calculate ETOTAL see Appendix E. the value of ImMAX. When connecting panels in parallel.uk/). the required current ImMAX is calculated as: ImMAX = 1. when the relation between the solar demanded energy and the energy available is greatest. for panels with an inclination of ß degrees. The Im(WORST MONTH) will be the largest value of Im. we calculate: • The current that the array of panels needs to provide. in Watts. With the data from the “worst month”.solardesign. which will determine the required number of batteries.21 Im (WORST MONTH) Once you have determined the worst month. Due to losses in the regulator and batteries. we can obtain the desired voltage and current.. while the current remains unchanged. the currents are .

Npp = ImMAX / IPmax The total number of panels is the result of multiplying the number of panels in series (to set the voltage) by the number of panels in parallel (to set the current). The necessary capacity depends on the energy available during the "worst month" and the desired number of days of autonomy (N). you need to connect several panels in series to reach your desired voltage. to use panels of nearly identical characteristics when building an array. Remember that you need to provide a few volts more than the nominal voltage of the battery in order to charge it. we first calculate the required energy capacity of our system (necessary capacity. Nps = VN / VPmax In order to calculate the number of panels in parallel (Npp). To estimate the capacity of our battery. To calculate the size of the battery we need to consider the maximum depth of discharge (DoD) that the battery allows: CNOM(Ah) = CNEC(Ah) / DoDMAX . It is very important. 24 or 48 V). CNEC). If it is not possible to find a single panel that satisfies your requirements. CNEC (Ah)= ETOTAL(WORST MONTH)(Wh) / VN(V) x N The nominal capacity of the battery CNOM needs to be bigger than the CNEC as we cannot fully discharge a battery. rounded up to the nearest integer. NTOTAL = Nps x Npp Capacity of the battery or accumulator The battery determines the overall voltage of the system and needs to have enough capacity to provide energy to the load when there is not enough solar radiation.244 Chapter 7: Solar Power summed together while the voltage remains unchanged. rounded up to the nearest integer. The number of panels in series Nps is equal to the nominal voltage of the system divided by the voltage of a single panel. You should try to acquire panels with VPmax a bit bigger than the nominal voltage of the system (12. you need to divide the ImMAX by the current of a single panel at the point of maximum power Ipmax.

The goal is to minimize voltage drops. and batteries. it is important to consider not the peak power of all your equipment. a regulator needs to be able to operate with a current ImaxReg at least 20% greater than the maximum intensity that is provided by the array of panels: ImaxReg = 1. keep in mind that the performance of the inverter varies according to the amount of requested power. You should try to minimize the length of the cables between the regulator. you will need to buy a new regulator with a larger working current. The thickness is chosen is based on the length of the cable and the maximum current it must carry. For security reasons. If your regulator does not support the current required by your system. When choosing an inverter. In order to calculate the thickness S of the cable it is necessary to know: . we divide the nominal voltage of our installation (VN) by the nominal voltage of a single battery (VNBat): Nbs = VN / VNBat Regulator One important warning: always use regulators in series.2 Npp IPMax DC/AC Inverter The total energy needed for the AC equipment is calculated including all the losses that are introduced by the DC/AC converter or inverter. but the peak power of the equipment that is expected to operate simultaneously. Using a 1500 Watt inverter to power a 25 Watt load is extremely inefficient. never in parallel. An inverter has better performance characteristics when operating close to its rated power. and type of regulators and inverters that you want to use. Using short cables will minimize lost power and cable costs. Cables Once you know the numbers of panels and batteries. The length depends on the location of your the installation.Chapter 7: Solar Power 245 In order to calculate the number of batteries in series (Nbs). In order to avoid this wasted energy. panels. it is necessary to calculate the length and the thickness of the cables needed to connect the components together.

01286 mm2/m). this is a minimum of 6 mm2.246 Chapter 7: Solar Power • The maximum current IMC that is going to circulate in the cable. For the other sections. In the battery-load subsystem it depends on the way that the loads are connected. Cost of a solar installation While solar energy itself is free. the equipment needed to turn it into useful electric energy is not. For security reasons that are some minimum values. S is chosen taking into consideration the cables available in the market. In the case of the panel-battery subsystem. 0. You should choose the immediately superior section to the one that is obtained from the formula. that minimum is 4 mm2. it is ImMAX calculated for every month. • The voltage drop (Va-Vb) that we consider acceptable in the cable. r is resistivity (intrinsic property of the material: for copper. The voltage drop that results of adding all possible individual drops is expressed as a percent of the nominal voltage of the installation. and L the length. but you must also replace and maintain various components of the system. The problem of equipment . for the cable that connects panels and battery. Typical maximum values are: Component Panel Array -> Battery Battery -> Converter Main Line Main Line (Illumination) Main Line (Equipment) Voltage Drop (% of VN) 1% 1% 3% 3% 5% Typical acceptable voltage drops in cables The section of the cable is determined by Ohm's Law: S(mm2) = r( mm2/m)L(m) ImMAX(A)/ (Va-Vb)(V) where S is the section. You not only need to buy equipment to transform the solar energy in electricity and store it for use.

In order to calculate the real cost of your installation. Our new table will look like this: Description # Unit Cost Subtotal Lifetime (Years) 20 5 10 5 Yearly Cost $60 $20 $2. For simplicity. and a solar system is implemented without a proper maintenance plan. In accounting terminology this is known as amortization.50 60W Solar panel 30A Regulator Wiring (meters) 50 Ah Deep cycle batteries 4 1 25 6 $300 $100 $1 / meter $150 $1. The first thing to do it is to calculate the initial investment costs. we do not include the costs of transport and installation but you should not overlook them. we include an illustrative example. Description 60W Solar panel (about $4 / W) 30A Regulator Wiring (meters) 50 Ah Deep cycle batteries Number 4 1 25 6 Unit Cost $300 $100 $1 / meter $150 Subtotal $1.50 $180 $262.225 The calculation of our investment cost is relatively easy once the system has been dimensioned. To figure out how much a system will really cost to operate we must estimate how long each part will last and how often you must replace it.200 $100 $25 $900 $2.225 Total: Annual Cost: .200 $100 $25 $900 Total: $2.Chapter 7: Solar Power 247 replacement is often overlooked. You just need to add the price for each piece equipment and the labor cost to install and wire the equipments together.

an annual cost of $262. The annual cost is an estimation of the required capital per year to replace the system components once they reach the end of their useful life.248 Chapter 7: Solar Power As you see. once the first investment has been done.50 is expected. .

Power needs to be provided. Each time you pierce a watertight enclosure. ice. it has to be protected from the rain. and other harsh elements. equipment needs power to work. Without proper grounding. A NEMA 4X or NEMA 6 provides excellent protection. the International Electrotechnical Commission (IEC) assigns an ingress protection (IP) rating. The National Electrical Manufacturers Association (NEMA) provides guidelines for protection of electrical equipment from rain. fluctuating mains power. This chapter will give you some idea of the practical problems you will be up against when installing wireless equipment outdoors. and even a light winds in the proper climate can annihilate your wireless links. Of course. For fixtures that pierce the body of an enclosure (such as cable glands and bulkhead connectors). nearby lightning strikes. dust.8 Building an Outdoor Node There are many practical considerations when installing electronic equipment outdoors. and other contaminants. you provide another potential place for water to seep in. An enclosure with a rating of NEMA 3 or better is suitable for outdoor use in a fair climate. A good outdoor enclosure should also provide UV protec249 . and the antenna should be mounted at a sufficient height. even from hose driven water and ice. wind. and will likely need to connect to an antenna and Ethernet cable. Waterproof enclosures Suitable waterproof enclosures come in many varieties. Metal or plastic may be used to create a watertight container for outdoor embedded equipment. Obviously. sun. An ingress protection rating of IP66 or IP67 will protect these holes from very strong jets of water.

When piercing an enclosure. UV stabilized silicone compound or other sealant can be used for temporary installations. If your wireless router or CPE includes support for 802. locally available parts can be repurposed for use as enclosures. are inserted between Ethernet switches and the device to be powered. Only use electronic components that are designed to be used in an embedded environment. finding NEMA or IEC rated enclosures may be a challenge in your local area. Mounting the box in the shade. These injectors supply power on the unused pairs. Before putting any piece of electronics in a sealed box. Nearly 13 Watts of power can be provided safely on a CAT5 cable without interfering with data transmissions on the same wire. The 802. use quality gaskets or o-rings along with a cable gland to seal the opening. and glued joints will eventually weaken and allow moisture to seep in. Other equipment. Newer 802. will add to the life span of the box as well as the equipment contained inside. DC power can be provided by simply poking a hole in your enclosure and running a wire. End span switches can supply power on the same wires that are used for data (pairs 12 and 3-6) or on the unused wires (pairs 4-5 and 7-8). either beneath existing equipment. But manufacturers are increasingly supporting a very handy feature that eliminates the need for an additional hole in the box: Power over Ethernet (POE). as well as to protect the equipment inside. Providing power Obviously. solar panel. and connecting mismatching gear can damage the injector and the equipment to be powered. some manufacturers (notably Cisco) disagree on power polarity. called mid span injectors. Often. remember that there will be no airflow. Of course. You can greatly extend the life of a plastic enclosure by providing some protection from the sun. If your motherboard requires a fan or large heat sink. electrical conduit housings.3af standard defines a method for supplying power to devices using the unused pairs in a standard Ethernet cable. Unfortunately. or even plastic food containers can be used in a pinch. or thin sheet of metal specifically for this purpose. and your electronics will likely bake to death on the tower.3af.250 Chapter 8: Building an Outdoor Node tion to prevent breakdown of the seal from exposure to the sun. you could in theory simply connect it to an injector. but remember that cables flex in the wind. an outdoor electrical box) you could even wire an AC outlet inside the box. be sure that it has minimal heat dissipation requirements. Rugged plastic or metal sprinkler boxes. If your enclosure is large enough (say. Read the fine .3af compliant Ethernet switches (called end span injectors) supply power directly to connected devices.

as the resistance of the small cable can present a fire hazard.net/~sfoskett/tech/poecalc. If your wireless equipment doesn t support power over Ethernet. A pair of passive POE devices can typically be purchased for under $20. and also hold it in place on windy days. These devices manually connect DC power to the unused wires on one end of the cable. crimp a CAT5 cable only using the data wires (pairs 1-2 and 3-6). equipment can be located inside a building. or simply build one yourself. Using a slotted or mesh parabolic. it is very important to use a stand off bracket. To make your own. When estimating wind resistance. Normal glass will introduce little attenuation. When mounting antennas on towers. plus enough to account for loss in the Ethernet run. and provide at least that much current and voltage. rather than a solid dish.html Once you know the proper power and electrical polarity needed to power your wireless gear. and not mount the antennas directly to the tower. you can still use the unused pairs in a CAT5 cable to carry power.gweep. but tinted glass will introduce unacceptable attenuation. and a matching barrel connector on the other. antenna alignment and protection. Stand off brackets need to be strong enough to support the weight of the antenna. power. you will need to find out how much power the device requires to operate. Be sure that the mounting brackets . Large antennas such as solid dishes or high gain sectorial panels can have considerable wind load. provided there is a window with ordinary glass through which the beam can travel. Mounting considerations In many cases. Remember. Here is an online calculator that will help you calculate the voltage drop for a given run of CAT5 : http://www. You can either use a passive POE injector. will help reduce the wind load without much affect on antenna gain. the total surface of the antenna structure must be considered. This greatly simplifies mounting. and connect the other end directly to a barrel connector inserted in the device s power receptacle.Chapter 8: Building an Outdoor Node 251 print and be sure that your injector and wireless equipment agree on which pins and polarity should be used for power. These brackets help with many functions including antenna separation. You don t want to supply too much power. as well as the distance from the center of the antenna to the point of attachment to the building. and can put a lot of force on to their mounts in strong winds. antennas can act like small sails. and weatherproofing problems. Then simply connect the transformer to pairs 4-5 (usually blue / blue-white) and 7-8 (brown / brown-white) on one end. but is obviously only useful in populated areas.

or your antennas will become misaligned over time (or worse. Secondly. If paint is chosen. it is important to be sure that the steel used is weatherproofed. This can be done using different lengths of steel. .252 Chapter 8: Building an Outdoor Node and supporting structure are solid. the pipe must also be vertical. Hot galvanizing is preferred. but not too much clearance that the antennas become too hard to reach if any service or maintenance is required. the standoff bracket will have to be designed to allow for this. As the equipment will be outside for all of its service life. or by using combinations of threaded rod and steel plates. it will be important to plan a yearly inspection of the mount and repaint when necessary. Stainless steel often has too high a price tag for tower installations. If it is being mounted on a tapered tower. Painting all steel with a good rust paint will also work. but may not be available in some areas. This way the antenna can be pivoted on the pipe for aiming. Figure 8.1: An antenna with a standoff bracket being lifted onto a tower. The pipe on the standoff bracket that the antenna will be mounted on needs to be round. fall off the tower entirely!) Mounting brackets must have enough clearance from the tower to allow for aiming.

bolt it to the lower section of the pole. The pole (called a gin pole in the trade) can then be removed. as seen from the center of the tower. A rope passing through the pulley will facilitate the raising of the next section. if required. Figure 8. Self-supporting towers Self supporting towers are expensive but sometimes needed. Tighten the guy wires carefully. After the cantilever section becomes vertical. Chose the points so that the angles. This can be as simple as a heavy pole sunk into a concrete piling. while the two tower sections are attached with an articulated joint. When installing guyed towers. but for very tall structures a self supporting tower might be required. ensuring that you use the same tension at all suitable anchoring points. are as evenly spaced as possible.Chapter 8: Building an Outdoor Node 253 Guyed towers A climbable guyed tower is an excellent choice for many installations.2: A climbable guyed tower. or as complicated as a professional radio tower. . and the operation may be repeated. particularly when greater elevation is a requirement. a pulley attached to the top of a pole will facilitate the tower installation. The pole will be secured to the lower section already in place.

high powered FM may interfere with your wired Ethernet cable.254 Chapter 8: Building an Outdoor Node Figure 8. FM station antennas are acceptable. Whenever . provided that at least a few of meters of separation is kept between the antennas. although AM transmitting station antennas should be avoided because the whole structure is active. Be aware that while adjacent transmitting antennas may not interfere with your wireless connection. An existing tower can sometimes be used for subscribers.3: A simple self-supporting tower.

sandbags. avoiding potential leaks. Using such a rooftop “sled” eliminates the need to pierce the roof with mounting bolts. The base is then weighed down with bricks. water jugs. Figure 8. or just about anything heavy.4: A much more complicated tower. . These consist of a tripod mounted to a metal or wooden base.Chapter 8: Building an Outdoor Node 255 using a heavily populated antenna tower. Rooftop assemblies Non-penetrating roof mount antenna assemblies can be used on flat roofs. be very scrupulous about proper grounding and consider using shielded cable.

or water bottles to make a stable platform without penetrating a roof. rocks. Wall mount or metal strap assemblies can be used on existing structures such as chimneys or the sides of a buildings.256 Chapter 8: Building an Outdoor Node Figure 8. Use dielectric grease on the connection between two metals of different type to prevent any electrolysis effect. If the antennas have to be mounted more than about 4 meters above the rooftop. Copper should never touch galvanized material directly without proper joint protection. Dissimilar metals To minimize electrolytic corrosion when two different metals are in moist contact. their electrolytic potential should be as close as possible.5: This metal base can be weighed down with sandbags. Water shedding from the copper contains ions that will wash away . a climbable tower may be a better solution to allow easier access to the equipment and to prevent antenna movement during high winds.

Stainless steel can be used as a buffer material. when you can start on the problem again after a good night s sleep. .Chapter 8: Building an Outdoor Node 257 the galvanized (zinc) tower covering. the surface area of the contact should be large and the stainless steel should be thin. Many countries require special training for people to be allowed to work on towers above a certain height. remember that the tower will be there in the morning. hire a professional to do it for you. If you have never worked on a tower. and an over-tightened connector can break in extreme weather changes. Be sure to tighten connectors firmly. The sealant protects the connector from water seepage. Cables should have an extra drip loop to prevent water from getting inside the transceiver. Remember that it is extremely hazardous to work in the dark. If you run out of time. If it is used as a buffer between copper and galvanized metals. Avoid working on towers during strong winds or storms. then a layer of sealing tape. connectors should be protected by applying a layer of electrical tape.6: A drip loop forces rainwater away from your connectors. Give yourself plenty of time to complete the job long before the sun sets. and then another layer of electrical tape on top. Joint compound should also be used to cover the connection so water can not bridge between the dissimilar metals. and the tape layer protects the sealant from ultraviolet (UV) damage. but never use a wrench or other tool to do so. Remember that metals expand and contract as temperature changes. Tower work will likely take longer than you think it will. To Access Point To Antenna Water flow Water flow Figure 8. and only when there is plenty of light. Safety Always use a harness securely attached to the tower when working at heights. Always climb with a partner. Protecting microwave connectors Moisture leakage in connectors is likely the most observed cause of radio link failure. but you should be aware that stainless steel is not a very good conductor. Once tight.

.4 GHz. For example. since they often include an external SMA antenna connector and can be powered by a small battery. This can be very difficult to find with a spectrum analyzer. The ideal antenna alignment toolkit consists of a signal generator and a spectrum analyzer.258 Chapter 8: Building an Outdoor Node Aligning antennas on a long distance link To properly align antennas at a great distance. The use of a signal generator is preferable to using the radio card itself.4 GHz to 5 GHz converter. A WiFi card transmits many discrete packets of information. they permit you to test the low.4 GHz ISM band. While these do not directly correspond to WiFi channels. you can observe the received power and watch the effect of moving the antenna to various positions in real time. ultimately stopping when the maximum received power has been found. This lets you to make small changes to the antenna alignment while watching the feedback tool. Fortunately there are a number of inexpensive tools that can be used instead. Once the maximum has been found on one end of a point to point link. the generator and analyzer can be swapped. Obviously. These devices accept a low power 2. preferably one of each at both ends of the link. Television transmitters (sometimes called video senders) are particularly useful.4 GHz (or even 5 GHz if using 802. you will need some sort of visual feedback that shows you the instantaneous received power at the antenna feed. the cost of a calibrated signal generator and spectrum analyzer that works at 2.11a) is well beyond the budget of most projects. as the signal generator can generate a continuous carrier. you can use a video sender in combination with a 2. middle. baby monitors. Video senders usually include support for three or four channels. particularly when operating in noisy areas. Inexpensive signal generator There are many inexpensive transmitters that use the 2. or high end of the band. By attaching a signal generator to one end of the link and a spectrum analyzer to the other. and miniature television transmitters all generate a continuous signal at 2. and the process repeated for the other end. They are usually quite expensive ($300$500 each) but will still likely be cheaper than a 5 GHz signal generator and spectrum analyzer. For 5 GHz work. switching the transmitter on and off very rapidly.4 GHz signal and emit high power 5 GHz signals. cordless phones.

4 GHz spectrum analyzers is slowly coming down. Whatever you choose for a signal source. they still typically cost a few thousand dollars.7: A 2. It features a very sensitive receiver in a small form factor (about the size of a USB thumb drive). Figure 8. While the cost of 2. Wi-Spy The Wi-Spy is a USB spectrum analysis tool made by MetaGeek (http://www.4 GHz video sender with an SMA antenna connector.8: The Wi-Spy USB spectrum analyzer . you will need a way to display the received power level levels at the other end. even for used equipment.metageek.Chapter 8: Building an Outdoor Node 259 Figure 8.net/).

10: EaKiu's 3D view lets you rotate and zoom in on any part of the graph in real time.cookwareinc. Figure 8. There is probably a WiFi network on channel 11. maximum. Figure 8. In addition to the standard views. and adds support for multiple Wi-Spy devices. it also provides an animated 3D view.260 Chapter 8: Building an Outdoor Node The latest version of the Wi-Spy includes better dynamic range and an external antenna connector. It also comes with very good spectrum analysis software for Windows called Chanalyzer.com/EaKiu/). . There is an excellent free software package for Mac OS X called EaKiu (http://www. and spectral views.4 GHz television transmitter. It provides instantaneous.9: The distinctive spiked pattern to the left of the graph was caused by a high power 2. average. topographic. with other noise sources lower down in the band.

Ideally. Note that this will only work on the client end of a link. This package includes command line tools as well as a GUI built on GTK. then the process will take all day and will probably end with misaligned antennas. do clear. sleep 1. iwspy. With some routers. This causes the router to emit a loud tone. you can also enable an audio feedback mode. each team should have at least two people: one to take signal readings and communicate with the remote end. the Wi-Spy is supported by the Kismet Spectrum-Tools project (http://kismetwireless. done This will show the state of all radio cards in the system. iwconfig. you should use the iwspy command to collect statistics for the MAC address of the client: wildnet:~# iwspy ath0 00:15:6D:63:6C:3C wildnet:~# iwspy ath0 Statistics collected: 00:15:6D:63:6C:3C : Quality=21/94 Signal=-74 dBm Noise=-95 dBm Link/Cell/AP : Quality=19/94 Signal=-76 dBm Noise=-95 dBm Typical/Reference : Quality:0 Signal level:0 Noise level:0 You can then use a while loop (as in the previous example) to continually update the link status. Other methods Some wireless routers (such as the Mikrotik) provide an "antenna alignment tool" that shows you a moving bar representing the received power. On the access point (master mode) side. you will need to use the operating system to provide feedback about the wireless link quality. Keep these points in mind while working on long distance links. updating once every second. wildnet:~# while :. do clear. the antenna is aligned. You will have two teams of people.net/spectools/). If you change too many variables at once (say. one team starts wiggling an antenna while the other tries to take a signal strength reading). One simple method to do this in Linux is with a loop that continually calls iwconfig. sleep 1.Chapter 8: Building an Outdoor Node 261 For Linux users. For example: wildnet:~# while :. If you don't have a spectrum analyzer. changing the pitch according to the received power. or a device that supports an antenna alignment mode. done Antenna alignment procedure The key to successfully aligning antennas on a very long distance link is communication. When the bar is at the maximum. . Wi-Spy. the other to manipulate the antenna.

You don t necessarily need a GPS to get an idea of where to point your antenna. You should be able to return to this known good state by simply powering on the device. Test signal in both directions. Once both ends have made their best guess. Using a good monitoring tool (such as Kismet. or a good built-in wireless client). If you are building a link in an area with few landmarks.262 Chapter 8: Building an Outdoor Node 1. 6. build your own landmark. power everything on. without having to log in or change any settings. Take some time to document the location of each site. Now is a good time to agree on antenna polarization (see Chapter 2 if you don t understand what polarization means). While mobile phones are usually good enough for working in cities. You don t want to fiddle with settings once you re in the field. connect every antenna and pigtail. log the GPS coordinates and elevation as well. all the better. or if your teams have amateur radio licenses. flare. Netstumbler. This can be very useful later to determine the feasibility of another link to the location without having to travel there in person. If this is your first trip to the site. If you can use binoculars to see the other end. Once you have made your guess. . lock the antenna firmly into place and signal the other team to begin slowly sweeping around. Before separating the equipment. mobile reception can be bad or nonexistent in rural areas. After the best possible position is found. and make sure you can establish a connection between the devices. 4. balloon. Once the best position is found. Test all equipment ahead of time. but only one at a time. try altering the elevation of the antenna. Some kinds of terrain make it difficult to judge the location of the other end of a link. Use a compass to roughly align the antenna to the desired bearing. Large landmarks are also useful for pointing. you may already have signal. Bring backup communications gear. the end with the lowest gain antenna should make fix their antenna into position. or even smoke signal might help. If all else fails. including surrounding landmarks and obstructions. Repeat this process a couple of times until the best possible position for both antennas is found. To begin. Start by estimating the proper bearing and elevation. a self-made landmark such as a kite. 5. 3. Working at a distance can be very frustrating if you are constantly asking the other team “can you hear me now?” Pick your communication channels and test your radios (including the batteries) before separating. Bring a high powered FRS or GMRS radio. flood light. Bring a camera. 2. the team with the highest gain antenna should slowly sweep it horizontally while watching the signal meter. use a ham rig. both teams should use triangulation (using GPS coordinates or a map) to get a rough idea of the direction to point. take a signal strength reading. If you are close enough and have made a good guess.

it is always prudent to use circuit breakers instead. The same goes for the team on the other side of the link. or even directly at the ground. If done properly. If nothing works. 9. radiation patterns incorporate many smaller sidelobes of sensitivity. Too often. with no damaged or suspect parts? As outlined in chapter eight. 8. it should be fun! Surge and lightning protection Power is the greatest challenge for most installations in the developing world. they are often poorly controlled. when taking signal strength readings. It can be frustrating to attempt aligning a dish only to discover that the other team is using the opposite polarization. but all of the equipment connected to it. By working methodically and communicating well. a double check doesn t hurt. If your received signal is mysteriously small. In rural areas. Work slowly and communicate your status well with the other team. in addition to a much larger main lobe. Continue sweeping slowly beyond that lobe to see if you can find the main lobe. Don t touch the antenna when taking a reading. Are the devices on both ends of the link powered on? Are all pigtails and connectors properly connected. Proper surge protection is critical to not only protect your wireless equipment. The antenna angle may look completely wrong. you may have found a sidelobe. you are concerned with finding the best possible position to achieve the greatest possible received signal. and don t stand in the path of the shot. These may need to be imported. but if a link stays stubbornly weak. but very often neglected. this should be agreed upon before leaving home base. Do not touch the antenna. replaceable . but shouldn't be overlooked. Fuses and circuit breakers Fuses are critical. too. fluctuate dramatically and are susceptible to lightning.Chapter 8: Building an Outdoor Node 263 7. Offset feed dishes will seem to be pointing too far down. As we saw in chapter four. 11. Despite the added cost. Your body will affect the radiation pattern of the antenna. check all components one at a time. The main lobe of an antenna often radiates slightly to one side or the other of the visual dead center of the antenna. 10. Again. Don t be afraid to push past the best received signal. Where there are electrical networks. and even in many urban areas of developing countries. proper troubleshooting technique will save you time and frustration. fuses are difficult to find. you can complete the job of aligning high gain antennas in just a short while. Don t worry about how the antenna looks. Double-check polarization.

pipes aren t yet in the ground. Static can cause significant degradation to signal quality. particularly on sensitive receivers (VSATs for example). Underground copper pipes can be very well grounded (depending on their depth. you are trying to accomplish two things: provide a short-circuit for a lightning strike. An effective ground will dissipate small amounts of energy to the ground. Take a light socket and a low-wattage bulb (30 Watts). all of the electronic equipment at at rural radio station was destroyed when a lightning strike went through the circuit. To ground the equipment.264 Chapter 8: Building an Outdoor Node fuses are removed and pocket change is used instead. salinity. Providing the short-circuit is simple. the moisture. The installer simply needs to make the shortest path from the highest conductive surface (a lightning rod) to the ground.e. When grounding. 3. When a strike hits the rod. If your ground is not efficient you will need to bury a grounding stake deeper (where the soil is more moist. while the second provides a path to dissipate excess energy that would otherwise cause a build-up of static electricity. amount of metal and organic content of the soil). If the ground is working. In a recent case. tropical soils). The more sophisticated way is to simply measure the impedance between the positive circuit and the ground. Then use a thick gauge conductive wire to connect the rod to something that itself is well grounded. the bulb should shine slightly. like 8 gauge braided copper). This LED is lit by energy that is being diffused to the ground circuit. The least accurate is to simply plug a good quality UPS or power strip into the circuit that has a ground detect indicator (a LED light). A common approach where there . mount a lightning rod above the equipment on a tower or other structure. has more organic matter and metals) or you need to make the ground more conductive. and provide a circuit for excess energy to be dissipated. you need thick gauge wire. and previous grounding equipment is often inadequate due to ill-conductive soil (typical of seasonally arid. In many sites in West Africa. as this energy does not turn an electrical counter! 2. This ground should be able to handle high-voltage (i. the energy will travel the shortest path and thus by-pass the equipment. The first step is to protect equipment from a direct or near direct lightning hit. without circuit breaker or even a fuse to protect it. connect one wire to the ground wire and the second to the hot wire. There are three easy ways to measure the efficiency of your ground: 1. How to ground Proper grounding doesn t have to be a complicated job. Some people actually use this to pirate a bit of free light.

such as an iron anvil or steel wheel. but most are either digital or electromechanical. This is normally effective. then attach a ground cable from that point. They often burn out after just one strike. which always produces the desired voltage (normally 220V). use a digital regulator. and will offer better protection for the rest of your equipment. Once burnt. Be sure to inspect all components of your power system (including the stabilizer) after lightning activity. . Whenever possible. The latter are much cheaper and more common. it too can be used to ground the tower. Soak the area. They are more expensive. and the charcoal and salt will diffuse around the hole and make a conductive area surrounding your plomb. This is sometimes called a plomb. simply peel back a bit of cable at the point closest to the ground before it goes into the building. Power stabilizers & regulators There are many brands of power stabilizers. Then fill the hole with charcoal and mix in salt. 240V. but these units offer little protection from lightning or other heavy surges. they can actually be fused at a certain (usually wrong) output voltage. This then needs to be waterproofed. Drop in a highly conductive piece of metal that has some mass to it. improving the efficiency of the ground. To ground the cable.Chapter 8: Building an Outdoor Node 265 is little soil is to dig a hole that is 1 meter in diameter and 2 meters deep. or 110V and use that energy to turn a motor. If radio cable is being used. though a more resilient design is to separate the ground for the tower from the cable. They are worth the added cost. either by soldering or using a very conductive connector. then top with soil. but are much less susceptible to being burnt. which literally means lead but can be any heavy piece of metal weighing 50 kg or more. Digital regulators regulate the energy using resistors and other solid state components. Electromechanical stabilizers take power at 220V.

.

Even though these indi267 . This chapter details a series of strategies to help you build a team that can support your network effectively. where an older individual is likely to be more capable of making decisions that positively affect the network as a whole. Though these people are invaluable resources. they take the knowledge with them when they go. There may also be many young and ambitious teenagers or young adults who will be interested and have the time to listen. they are very helpful and will learn quickly. These people will take interest in your network and want to learn as much about it as possible. re-wiring a broken television or welding a new piece to a bicycle. and new physical obstructions can cause a long-running network to fail. Unlike wired connections. wind. problems with a wireless network are often invisible. Interference. especially the ambitious youth who tend to want to be involved. company or family has individuals who are intrigued by technology. Again. but the project team must focus their attention on those who are best placed to support the network in the coming months and years. help. you must avoid imparting all of the specialized knowledge of wireless networking to only one person. They are the ones found splicing the television cable.9 Troubleshooting How you establish the support infrastructure for your network is as important as what type of equipment you use. If your only specialist loses interest or finds better paying work somewhere else. Young adults and teenagers will go off to university or find employment. Building your team Every village. and learn about the network. and can require more skill and more time to diagnose and remedy. These youth also have little influence in the community.

The next strategy was to divide functions. and to document all methodology and procedures. that is not often the best approach. and to train people who were permanently rooted in the community: people who had houses and children. He was very motivated and learned quickly. All of the knowledge about the network and how to support it left with him. Values. If they are well educated in this process. who can be motivated. They will be best able to judge who will work best with them. thus compromising their ability to manage. The first time you look at an abstract painting. but it is best to provide a means of checks and balances. and he was able to deal with a variety of problems. while remaining completely unfathomable to people who are not from that community. make sure that they understand that criteria. A complementary strategy is to compartmentalize functions and duties. he was taught more than had been foreseen. and were already employed. It is often best to find a local partner organization or a local manager. It is also important that the local partner is given this authority and is not undermined by the project organizers. then your requirements should be satisfied. and to set firm boundaries. Where a candidate is kin. Even a better salary could not keep him. Therefore. For example.268 Chapter 9: Troubleshooting viduals might have less time to learn and might appear to be less interested. Troubleshooting and support of technology is an abstract art. people can be trained easily. to provide them sound criteria. and many other factors will be important to them. Because he learned so quickly. In this way. but the community will retain this knowledge for much longer. and substituted with little effort. The best approach is to coach your local partner. from fixing a PC to rewiring Ethernet cable. in one project site the training team selected a bright young university graduate who had returned to his village. since the prospect of a stable government job was too appealing. Unfortunately. their involvement and proper education about the system can be critical. and work with them to find the right technical team. and teach them. though these rules must consider the local situation. who have roots in the community. The training team had to return and begin the training again. but do not let them capitalize use or knowledge of these systems. local politics. history. You should involve the youth. it may just look to you like a bunch of random paint . Find people who are committed to the community. It may be impossible to say that you cannot hire kin. It took three times as long to teach three people as it took to train the young university grad. Though this might seem to suggest that you should hand-pick who is to be involved. two months after the project launch he was offered a government job and left the community. there should be clear criteria and a second authority in deciding upon their candidacy. Such boundaries should include rules about nepotism and patronage. a key strategy in building a support team is to balance and to distribute the knowledge among those who are best placed to support the network for the long term.

you might suspect a power supply related problem and thus skip most of the procedure. these charts should not be followed blindly. Give them the cable crimper or keyboard and show them how to do the work. With those possibilities in mind. or orient the antenna in the wrong direction. have them apply this procedure to a more complicated problem. but it can take a while for them to appreciate the point of the “invisible” network. You then test to see if the experiment yields something similar to the expected result. Insert a broken fuse. but likely will not have had the opportunity to troubleshoot complex problems. if a long-running access point suddenly stops working after a storm. you complete the experiment. as there will be times when the problem is neither known nor evident. and the “invisible” coherence becomes very real. In rural areas. Test the student. Once the participants are chosen and committed to the project. use the wrong SSID. For example. Disconnect the aerial. making it clear that each problem will show specific symptoms. they might not think to apply it to resolving real problems. Of course. The typical agrarian villager may have been introduced to the concept. and try to eliminate the variables to isolate the problem. a phased approach is needed to ease people into supporting technology systems. The neophyte looking at a wireless network may see the antennas and wires and computers. you can change an IP address. Once they have fixed the television. Simple decision trees or flow charts can be made that test these variables. Even if you do not have time to explain every detail and even if it will take longer. you may come to appreciate the work as a whole. although time consuming. you take a set of variables. Therefore.Chapter 9: Troubleshooting 269 splatters. It is important that they develop a methodology and procedure to resolve these problems. it can often take a huge leap of understanding before locals will appreciate an invisible network that is simply dropped into their village. Simply put. The scientific method is taught in virtually all western schools. have your student develop a problem resolution procedure on something simple and familiar. It can be sped up by making logical assumptions. Let them "drive". It is often easier to teach this method using a non technological problem first. Give them a battery that is not charged. Start by sabotaging the television. People charged with supporting technology should be taught how to troubleshoot using this method. and point the way as to how to proceed. they need to be involved physically and see not only what has been done. In a network. If it did not. switch or damage cables. but how much work was done. Even if they are familiar with the scientific method. Many people learn about it by the time they reach high-school science class. . you recalculate your expected result and try again. After reflecting on the composition for a time. The best method is involvement. involve them as much as possible. This method is very effective. then slowly eliminate those variables through binary tests until you are left with one or only a few possibilities. For example. like a battery powered television.

having a backup means that you can quickly restore it to the previous settings and start again. • What was the last thing changed? If you are the only person with access to the system. This is much faster and less errorprone than re-crimping a cable. • Don t panic. • Is it plugged in? This step is often overlooked until many other avenues are explored. • Make a backup. survey the scene and assess exactly what is broken. This applies before you notice problems. antenna cable. saving your work at a given point lets you return to it as a known good. If you have historical logs or statistics to work from. what is the last change you made? If others have access to it. you may carry a tested Ethernet cable in a tool kit. For example. Trust me. or a CD-ROM with a known good configuration for the system. Plugs can be accidentally (or intentionally) unplugged very easily. that means that it was working at one time. what is the last change they made and when? When was the last time the system worked? Often. but you will feel even sillier if you spend a lot of time checking out an antenna feed line only to realize that the AP was unplugged the entire time. A known good is any component that you can replace in a complex system to verify that its counterpart is in good. probably very recently. all the better. as well as software. so you can make an informed decision before making changes. This idea applies to hardware. Roll back that change and see what effect it has on the problem. you can easily swap out the suspect cable with the known good and see if things improve. When troubleshooting very complex problems. system changes have unintended consequences that may not be immediately noticed. as well as after. you may also pack a backup battery. having a configuration that “sort-of” works can be much better than having a mess that doesn t work at all (and that you can t easily restore from memory). If you suspect problems with a cable in the field. Is the lead connected to a good power source? Is the other end connected to your device? Is the power light on? It may sound silly. . Be sure to collect information first. it happens more often than most of us would care to admit.270 Chapter 9: Troubleshooting Proper troubleshooting technique No troubleshooting methodology can completely cover all problems you will encounter when working with wireless networks. Before jumping in and making changes. • The known good. But often. even if the problem is not yet completely solved. and immediately tells you if the change fixes the problem. Here are a few simple points to keep in mind that can get your troubleshooting effort working in the right direction. When fixing complicated problems. If you make a complicated software change to a system. If you are troubleshooting a system. working condition. Likewise. problems come down to one of a few common mistakes.

but lead to more unintended consequences that break other parts of the system. or arrange to have another backup sent in. This could mean having an access point or router pre-configured and sitting in a locked cabinet. your changes may fix the original problem. plainly labeled and stored with backup cables and power supplies. These threats can range from the benign to the outright malevolent. When under pressure to get a failed system back online. then you will not understand exactly what led to the problem in the first place. It is unlikely that the people who design your network will be on call twentyfour hours per day to fix problems when they arise. This includes automated access from search engines that periodically spider your entire site. If you are not sure if a particular change will damage another part of the system. Your troubleshooting team will need to have good troubleshooting skills. but all will have impact on your network if it is not properly configured. One solution to this problem is to use split DNS and mirroring. it is tempting to jump ahead and change many likely variables at once. If you do. and either send the broken part to an expert for repair. Worse. and your changes seem to fix the problem. connectivity problems come from failed components. By changing your variables one at a time. It is often much more efficient to have a number of backup components on-hand. and be able to see the direct effects of the changes you make. Common network problems Often. Locally hosted websites If a university hosts its website locally. or simple misconfiguration. don t be afraid to call in an expert. Putting a penny in place of a fuse may solve the immediate problem. • Do no harm. but may not be competent enough to configure a router from scratch or crimp a piece of LMR-400. Assuming that the backups are kept secure and are replaced when used. Once your network is connected to the Internet or opened up to the general public. then either find someone with more experience or devise a way to test your change without doing damage. and train your team to be able to swap out the entire broken part. considerable threats will come from the network users themselves. adverse weather. you can precisely understand what went wrong in the first place. Your team can swap out the failed component. this can save a lot of time for everyone. but it may also burn down the building. This section looks at some common problems found once your network is used by actual human beings. The university mirrors a copy of its websites to a . If you don t fully understand how a system works.Chapter 9: Troubleshooting 271 • Change one variable at a time. visitors to the website from outside the campus and the rest of the world will compete with the university's staff for Internet bandwidth.

if your proxy server lies behind a border firewall. Internet VSAT Public Web Server Example 1 Internet VSAT Internal Web Server Example 2 Mirrored Public Web Server Figure 9. you can specify the IP address range of the campus network in your squid. For example. . This is because people elsewhere will connect and use open proxies for a variety of reasons.conf file as the only network that can use Squid. such as to avoid paying for international bandwidth.272 Chapter 9: Troubleshooting server at. while users on the university network access the same site locally. say. In Example 2. while a copy is kept on an internal server for very fast local access. not from the rest of the Internet. This improves the VSAT connection and reduces load times for web site users. all website traffic coming from the Internet must traverse the VSAT. and uses split DNS to direct all users from outside the university network to the mirror site. Open proxies A proxy server should be configured to accept only connections from the university network.1: In Example 1. Alternatively. you can configure the firewall to only allow internal hosts to connect to the proxy port. a European hosting company. The way to configure this depends on the proxy server you are using. Details about how to set this up are provided in chapter three. the public web site is hosted on a fast European service.

the reply after the first MAIL FROM should be something like: 550 Relaying is prohibited. An online tester is available at sites such as http://www. if an interactive command-line conversation can take place (for example. WinMX and BearShare can be prevented in the following ways: • Make it impossible to install new programs on campus computers.com Instead..rcpt to spammer@waste.uzz.com 250 OK . This will likely be the IP address range of the campus network. A disk . an institution that does not protect its mail systems is almost guaranteed to be found and abused. Sendmail. it may be necessary to type 'set local_echo' before the text is visible): telnet mail. Morpheus. BitTorrent. it is possible to prevent the installation of programs such as Kazaa.Chapter 9: Troubleshooting 273 Open relay hosts An incorrectly configured mail server will be found by unscrupulous people on the Internet.com> RCPT TO: innocent@university.org/. The PC is also configured in a way that prevents users from installing new applications. the server is an open relay host: MAIL FROM: spammer@waste. Peer-to-peer networking Bandwidth abuse through peer-to-peer (P2P) file-sharing programs such as Kazaa.zz 250 OK . and avoid getting caught.ac. They then install all the necessary applications on it. To test for an open relay host. They do this to hide the true source of the spam. and configure these in an optimal way. Postfix. Use telnet to open a connection to port 25 of the server in question (with some Windows versions of telnet. By not giving regular users administrative access to PC workstations. where they install the required operating system on one PC.zz 25 Then.mail from <spammer@waste. the following test should be carried out on your mail server (or on the SMTP server that acts as a relay host on the perimeter of the campus network). Exim. Configuring the mail server not to be an open relay consists of specifying the networks and hosts that are allowed to relay mail through them in the MTA (eg. and be used as a relay host to send bulk email and spam. Since bulk emailers have automated methods to find such open relay hosts. as follows). Many institutions also standardize on a desktop build. or Exchange).ordb.ac. There is also information about the problem at this site.

Network administrators simply cannot predict how users will make innovative use of a network. causing the operating system and all software on the computer to be exactly as specified.partimage. and all other services not specifically permitted by the proxy server. Some programs are spyware. you will prevent users from making use of any services (even low-bandwidth services) that your proxy does not support. For this reason. Kazaa defaults to port 1214 for the initial connection. From time to time. If these are blocked. This is because Kazaa and other protocols are clever enough to bypass blocked ports. using bandwidth management tools.for example. such as Spychecker (http://www. for example). Programs that install themselves (from the Internet) There are programs that automatically install themselves and then keep on using bandwidth . By preemptively blocking all access.de/). making it look like web traffic. the so-called Bonzi-Buddy. this would block all P2P traffic. VPN software. there are software solutions to find and remove these problem programs. .powerquest. • Blocking these protocols is not a solution. ISPs don't block it but "throttle it".274 Chapter 9: Troubleshooting image of this PC is then cloned to all other PCs using software such as Partition Image (see http://www. but shouldn t be taken lightly. When this happens. While this may be desirable in extremely low bandwidth circumstances. It would also block all other types of traffic. • If rate-limiting is not an option.org/) or Drive Image Pro (see http://www. These programs are preventable to some extent by user education and locking down PCs to prevent administrative access for normal users. users may succeed in installing new software or otherwise damaging the software on the computer (causing it to hang often. If the proxy server and mail servers are configured with two network cards (as described in chapter three) and these servers are not configured to forward any packets. and some kinds of worms. In other cases. Such a decision may be necessary. SSH. it should never be considered as a good access policy in the general case. an administrator can simply put the disk image back. such as Microsoft NetMeeting.com/).lavasoft. but if that is not available it will attempt to use ports 1000 to 4000. the Microsoft Network. which keep sending information about a user's browsing habits to a company somewhere on the Internet. In low bandwidth networks it may be decided that the simplicity of this design will outweigh the disadvantages.spychecker. change the network layout.com/) or Ad-Aware (http://www. its uses port 80.

cab *DENIED* Banned extension .cab HEAD 0 While this may be tolerable for a few PC clients.4. Unfortunately. This is a free program from Microsoft that enables you to download all the updates from Microsoft overnight on to a local server and distribute the updates to client workstations from there.com/ident.4.4.1.com/ident.cab GET 0 2003.cab GET 0 2003. the problem grows significantly as hosts are added to the network.168.Chapter 9: Troubleshooting 275 Windows updates The latest Microsoft Windows operating systems assume that a computer with a LAN connection has a good link to the Internet.21 http://windowsupdate. bug fixes and feature enhancements from the Microsoft Web site. This is only a good option for large networks.cab *DENIED* Banned extension .com/ident.168.1.cab GET 0 2003. but can save untold amounts of Internet bandwidth.microsoft.microsoft.cab *DENIED* Banned extension . and automatically downloads security patches.cab *DENIED* Banned extension .21 http://windowsupdate.cab *DENIED* Banned extension .microsoft.2 13:24:17 192.com/ident. it places a heavy load on the proxy server.1.168.21 http://windowsupdate. and if all workstations do that. In this way.2 13:24:18 192.4.cab GET 0 2003. Windows updates need not use any bandwidth on the Internet link during the day.21 http://windowsupdate.cab) files.microsoft.1.com/ident.microsoft.microsoft. Blocking the Windows updates site on the proxy server is not a good solution because the Windows update service (Automatic Updates) keeps retrying more aggressively.microsoft.cab *DENIED* Banned extension .168.21 http://windowsupdate.2 13:24:21 192. The two possible approaches to this problem are: • Disable Windows updates on all workstation PCs.4.microsoft.2 13:24:19 192.com/ident. The security updates are very important for servers.1.21 http://windowsupdate. The extract below is from the proxy log (Squid access log) where this was done by blocking Microsoft's cabinet (. Much of the Squid log looks like this: 2003. you can also configure it to answer requests for windowsupdate.com/ident.com/ident.168.cab GET 0 2003.1.microsoft.4.cab GET 0 2003.21 http://windowsupdate. This can consume massive amounts of bandwidth on an expensive Internet link.1.com/ident.21 http://windowsupdate. If you have a flexible DNS server.168.cab *DENIED* Banned extension .2 13:24:21 192.2 13:24:21 192.cab GET 0 2003.com and direct the updater to your update server.cab HEAD 0 2003. but whether workstations in a protected private network such as a campus network need them is debatable.1.cab *DENIED* Banned extension .2 13:24:18 192.2 13:24:20 192.168. Rather than forcing the proxy .2 13:24:19 192.cab *DENIED* Banned extension .4.1. all client PCs need to be configured to use the Software Update Server for this to have an effect.4.168.168.21 http://windowsupdate. • Install a Software Update Server.4.microsoft.

and will exploit these connections to gain greater access to your network. and worse. The latest versions of Windows and Mac OS X also have a time synchronization service. This keeps the computer clock accurate by connecting to time servers on the Internet. . as well as upload usage patterns back to a site on the Internet. poses a potential security risk. It is a protocol that works by holding elections to determine which computer will be the master browser.276 Chapter 9: Troubleshooting server to serve requests that will always fail. It is better if these updates are distributed from a local server. rather than to tie up the Internet link with these requests. The following steps might be taken to prevent this: • Block outgoing SMB/NetBIOS traffic on the perimeter router or firewall. Many Internet worms and penetration tools actively scan for open SMB shares. It is more efficient to install a local time server and distribute accurate time from there. These applications may need to be throttled or blocked altogether. shares and printers that you can see in Network Neighborhood or My Network Places. wasting the organization's bandwidth. anti-virus packages (such as Norton AntiVirus) periodically update themselves automatically and directly from the Internet. Innocuous looking applets (like Konfabulator and Dashboard widgets) continually poll Internet hosts for updated information. and therefore consume bandwidth for reasons the user might not predict. Programs that assume a high bandwidth link In addition to Windows updates. it will also tend to spread to the Internet link. Unless SMB traffic is filtered. The master browser is a computer that keeps a list of all the computers. automatically download updates and advertisements. Other programs. The SMB protocol is designed for LANs and causes problems when the Windows computer is connected to the Internet. Windows traffic on the Internet link Windows computers communicate with each other via NetBIOS and Server Message Block (SMB). many other programs and services assume that bandwidth is not a problem. These can be low bandwidth requests (like weather or news updates). Information about available shares are also broadcast at regular intervals. This traffic will eat up Internet bandwidth. These protocols work on top of TCP/IP or other transport protocols. such as the RealNetworks video player. it makes more sense to redirect the Software Update clients to a local update server. or very high bandwidth requests (such as webcams). For example.

A PC should not have shares unless it is a file server. In fact. the less there is to exploit. even though it is an old one. only the file server should have any shares. For example.com/. user education about executing attachments and responding to unsolicited email is essential. • Reduce network shares. but Windows Explorer does not. Ideally. The W32/ Opaserv worm. it should be a policy that no workstation or server should run unused services. All emails sent to her in her absence are still forwarded to her Yahoo account. Furthermore. An email loop is formed that might send hundreds of thousands of emails back and forth. The user goes on holiday. it starts bouncing the emails back to the university account. Internet Explorer needs to connect to the Internet. the fewer services a computer runs. which can grow to only 2 MB.softperfect. It is therefore essential that anti-virus protection is installed on all PCs. Administrators must also take care that they do not turn this feature off by mistake.com/) to easily identify all the shares in your network. and a server should not run unnecessary services either. For example. There are features of mail server programs that can recognize loops. Worms and viruses Worms and viruses can generate enormous amounts of traffic. or install an SMTP forwarder that modifies mail headers in such a way that the mail server does not recognize the mail loop. For example. It spreads through Windows shares and is detected by other people on the Internet because it attempts to spread further. These should be turned on by default. generating massive traffic and crashing mail servers. This should be disabled if that server has a different function. a user whose university account is configured to forward all mail to her Yahoo account. which immediately forwards it back to the Yahoo account.Chapter 9: Troubleshooting 277 • Install ZoneAlarm on all workstations (not the server). a single user making a mistake can cause a problem. Email forwarding loops Occasionally. for example. This program allows the user to determine which applications can make connections to the Internet and which ones cannot. ZoneAlarm can block Windows Explorer from doing so.zonelabs. . A free version can be found at http://www. You can use a tool such as SoftPerfect Network Scanner (from http://www. Windows and Unix servers typically run a web server service by default. When the Yahoo account becomes full. is still prevalent.

The solutions to this kind of problem lie in training. type this from the command prompt: ftp -s:c:\ftpscript. the specified file is queued for later download and the user notified by email when the download is complete. or download large files such as 650MB ISO images. Therefore. If the offline link is selected. Sending large files When users need to transfer large files to collaborators elsewhere on the Internet. This system would only work for users who are not impatient. In this way.txt): open ftp. an upload to a remote FTP server can be done using an FTP script file. Users accessing ftp:// URLs are served a directory listing in which each file has two links: one for normal downloading.cmd. 2000 and XP computers. In Windows. the command can be saved into a file such as transfer. and are familiar with what file sizes would be problematic for download during the working day. and scheduled to run at night using the Sched- .278 Chapter 9: Troubleshooting Large downloads A user may start several simultaneous downloads. as outlined in chapter six).zip quit To execute. and retrieves such files immediately when requested again. they should be shown how to schedule the upload. a single user can use up most of the bandwidth. which is a text file containing FTP commands. The download queue is sorted by file size.ed. and monitoring (including real-time monitoring. similar to the following (saved as c:\ftpscript. users requesting small files may receive them within minutes. offline downloading.ac. small files are downloaded first. • Another approach would be to create a web interface where users enter the URL of the file they want to download.txt On Windows NT. a system was implemented using URL redirection. This is then downloaded overnight using a cron job or scheduled task. sometimes even faster than an online download.uk gventer mysecretword delete data.zip binary put data. and the other for offline downloading. Offline downloading can be implemented in at least two ways: • At the University of Moratuwa. The system keeps a cache of recently downloaded files. As some bandwidth is allocated to this system even during peak hours.

This is what the University of Bristol has done with their FLUFF system. . a web front-end can be written for a local web server to accept a large file and place it in a download area. He can then give that URL to his local or international collaborators. It is a waste of bandwidth to send these via the Internet if the recipient is local. A file share should be created on the local Windows / Samba /web Novell server. Users sending each other files Users often need to send each other large files. A system like this can easily be implemented as a CGI script using Python and Apache. the user receives a URL for the file.Chapter 9: Troubleshooting 279 uled Tasks (Start Settings Control Panel the same can be achieved by using at or cron.uk/fluff/. and when they access that URL they can download it.ac. In Unix. After uploading it to the web server. The University offers a facility for the upload of large files (FLUFF) available from http://www. where a user can put the large file for others to access. The advantage of this approach is that users can give external users access to their files.bristol. Scheduled Tasks). whereas the file share method can work only for users within the campus network. These files can then be accessed by anyone who has been given their location. Alternatively.

.

Substantial progress in the use of wireless networks for rural communications has been accomplished over the past few years. this chapter will focus on documenting methods for building sustainable wireless networks and telecenters. which is a substantial improvement. There is little wired infrastructure beyond the principal cities. the Macedonia Connects project has now connected a majority of the country's schools to the Internet. there has been tremendous growth in Internet access across the developing world. The prohibitive cost of Internet connectivity in many developing countries imposes a substantial operating expense that makes these models sensitive to economic fluctuations and necessitates innovation to attain viability. There are now proven models for rural access using wireless. The models described here are smaller in scale and use affordable designs. Most developing world cities now have wireless or ADSL networks and fiber optic connections to the Internet. Long-distance links have been constructed. Based on the authors experiences and observations of existing networks. In Macedonia. outside urban areas. due in large part to technological breakthroughs. In the past decade. there have been fewer successes with the development of sustainable business models for wireless networks and telecenters. Therefore. as well as knowledge from entrepreneurial development best practices. high bandwidth designs are possible and secure means to access networks are available. In contrast. wireless remains one of the few choices for providing affordable Internet access. particularly for remote areas. Internet access is still a formidable challenge. Nevertheless. This book was written for those wishing to connect their communities. Our aim is to provide examples of how wireless networks can be designed to expand sustainable 281 .10 Economic Sustainability Achieving long-term sustainability is perhaps the most difficult goal when designing and operating wireless networks and telecenters in developing countries.

many people assume that there is one preferred business model that will work in every community of the developing world. First. This chapter is not meant to be a guide for managing sustainable wireless networks. Two common misconceptions must be dispelled. Thus. In this circumstance. other innovative models must be customized to fit the context of this particular community.282 Chapter 10: Economic Sustainability access where large telecommunications operators have not yet installed their networks into areas that would otherwise not be economically feasible by traditional models. town or village is different. Although this term generally means that a system is built to persist indefinitely. When determining and implementing the best model for a wireless network or telecenter. The first step involves documenting this vision with the input of your entire team or staff. another village nearby may not possess the same necessary qualities for this model to be sustainable. Also. it centers on a time period of five years – the period in which these ICT infrastructure and wireless technologies are expected to be useful. In practice. Although one model may work in one village. this “how-to” guide seeks to present an approach that will enable you to find the model that best fits your situation. the process is ongoing and iterative. In fact. the term sustainability will be used to encapsulate a system designed to persist for approximately five or more years. All of the steps are integrally connected to each other. the characteristics of a sustainable business model vary from community to community. However. Rather. There is no prescribed model that meets the needs of all areas in the developing world. and often you will revisit steps several times as you progress. instead of the horizon being indeterminate. Keep in mind that determining the best model is not a sequential process where each step is followed until completion. Each community. several key factors help to ensure its success. and the key to success is to find that one “eureka” solution. What is the purpose of the wireless network? Who does the net- . Create a Mission Statement What do you want to accomplish by setting up your network? It seems like a simple question. this is not the case. many wireless networks are installed without a clear vision of what they are doing and what they hope to accomplish in the future. Another misconception is that sustainability has the same definition for all people. The tools and information contained within this chapter will help people starting wireless networks in the developing world to ask the right questions and gather the necessary data to define the most appropriate components of their model. Despite the fact that some places may be similar in economic terms. this chapter focuses more on the discussion of the economic conditions (financial and managerial) than other aspects of sustainability.

It is important that every team member working to build the wireless network is included in the process of developing your mission. the needs of customers and the best way to satisfy those needs change rapidly. the development of your mission is an ongoing process. Potential users could consist of a wide variety of individuals and organizations that include. groups and organizations in the community that have a need for information and would benefit from the wireless network s offerings. you must conduct research to determine whether this first conception is aligned with the realities of your environment. identify the individuals.Chapter 10: Economic Sustainability 283 work seek to serve? What does the network do to address the community s needs and to create value? What are the principles that guide the network? A good mission statement expresses the purpose of your network in a concise. partners and donors. which helps create further buy-in. which will further your overall objectives. After defining the initial mission with your team. you must constantly modify the mission throughout the life-cycle of the wireless network. Evaluate the Demand for Potential Offerings The next step in deriving your business model involves assessing the community s demand for the network s products and services. but are not limited to: • Farmers associations and cooperatives • Women s groups • Schools and universities • Businesses and local entrepreneurs • Health clinics and hospitals • Religious groups • International and local non-governmental organizations (NGOs) • Local and national government agencies • Radio stations • Organizations in the tourist industry Once you establish a list of all the potential user groups of the network. It will garner support and commitment not only from your staff. A farmer may need to gather in- . Based on an analysis of the external environment and your internal competencies. meaningful way while articulating your values and services. you must determine their needs for access to information and communication. your mission provides a vision of the aspirations for your wireless network. Above all. people confuse services with needs. Often. but also from customers. In the dynamic world of technology. therefore. First.

however. you should connect with potential partners and explore mutually beneficial collaborations. it will ultimately fail. When the telecenter manager calculated the total monthly costs incurred by the NGO. The goal of this data collection is to obtain a thorough understanding of the demand for information and communication in your community so that the network being created responds to those needs. it is important to figure out where the network can bring the most value to its users. Mali. Your wireless network should look for the best way to fulfill the farmer s needs. focus groups. Establish Appropriate Incentives Often. wireless networks that do not succeed in the developing world forget this key step. the farmer could also receive this information through SMS over a mobile phone or through Voice over Internet Protocol (VOIP). industry reports. interviews or town hall meetings. magazines. thereby creating value at the lowest cost for the user.284 Chapter 10: Economic Sustainability formation on market prices and climatic conditions to improve his crop yield and sales. During this phase. and getting an Internet connection far outweighs the . as well as the opportunity cost of having the employee out of the office for several days each month. the NGO sent one of its employees to Mopti once a month. Conducting research through a review of statistical documentation. It is important to differentiate between needs and services because there may be various ways to satisfy the farmer s needs. he was able to demonstrate the value of an Internet connection through cost savings to the organization. a telecenter manager evaluated the potential benefits of establishing a wireless network through discussions with several local organizations. learning to use it. Often. When assessing the needs of the community. In addition. Your entire network should be based on the demand in the community. He interviewed one local NGO that discussed its need to send monthly reports to its headquarters office in Bamako. You can evaluate the demand in your community by contacting your potential customers and asking questions directly through surveys. there was no Internet access in Douentza. At that time. censuses. the cost of acquiring a computer. there is little economic incentive for such subsistence-based economic participants to access the Internet. If you set up a wireless network in which the community does not find value or cannot afford its services. In order to email a copy of the report. Assistance from key partners may also be necessary to secure sustainability for your wireless network. Perhaps the way in which he gets this information is through the Internet. For instance. resulting in transportation and lodging costs. in the small town of Douentza. newspapers and other secondary data sources will also help to give you a better picture of your local environment.

if providing market information for the local maize industry. How much perceivable economic value can be generated? 3. or terminals to sellers and merchants ($2/hr). Your network would then likely need to tie-into market information systems. quality standards imposed by importing countries. If the market was small. Additionally. It is crucial to design a network with viable economic uses and with costs that are less than the economic value provided by it. . You might also provide wireless connections to merchants and rent them thin-client terminals for Internet access. Your network might also provide the means for farmers to read about new techniques and to buy new products. The network must provide economic value to its users in a way that outweighs its costs. “By using this network you can improve your margins on commodity sales by 2%.” You must figure out how your network can improve efficiencies.Chapter 10: Economic Sustainability 285 economic returns that it can provide. you might be able to reduce costs by limiting access to images and other bandwidth intensive services. Internet access becomes an obvious advantage in situations where knowing the day-to-day prices of products can make a significant difference in income. Establishing appropriate economic incentives is paramount to the success of the network.” or “Internet will allow you to save $X in phone charges and transportation costs per month. the network will be able to clearly articulate its value proposition for its users. knowing how much value your network will create for these merchants will allow you to gauge how much they will be able to afford for your services. you must involve the community in the creation of the network from the beginning of the project. the network should be located near to where farmers bring their crop for sale to merchants. Again. and commodities exchanges. or it must be cheap enough that its costs are marginal and affordable to its users. reduce costs. making sure that this initiative is organic and not imposed from the outside. to create a proper incentive structure. or increase revenues for these customers. For example. you should try to answer the following questions: 1. For example. To begin. Can present impediments be overcome to allow the achievement of these economic returns? By answering these questions. There has recently been some development of applications that address this lack of incentive. What economic value can this network generate for the local economy and for whom? 2. providing daily price sheets ($1 each). such as market information systems.

.4 GHz frequencies. nothing would prevent you from offering the VoIP service. In Ukraine. reputation. In most situations. the government requires an expensive license to use 2.g. which renders this shared usage prohibitive. marketing. ensuring your network will comply with the laws of the country and local community. another wireless Internet service provider or WISP). In Syria. Some project managers have been forced to shut down their wireless networks simply because they were unknowingly breaking the law. Typically only well established Internet Service Providers in this country have sufficient cash flow to pay the license fees. surveys of their customers and visits to their site. and organizations defined as new entrants to the wireless market. Most countries in the developing world have not yet defined whether VoIP is permitted. Create a file for each competitor.4 GHz is free to use worldwide. Competitors include organizations that provide similar products and services (e. in some countries there are complicated rules surrounding VoIP. Although wireless networks are legal in the Ukraine. however. organizations viewed as substitutes or alternatives to the products and services your network provides (e. Because there are no such restrictions on wireless networks. the possibility to share Internet connectivity in small communities is a viable solution. research whether any organization has the right to use 2. You can obtain information about your competitors through the Internet. VoIP is legal for international calls only. you should research them thoroughly. a cybercafé).. You should also check into the legality of Voice over Internet Protocol (VoIP) services. such as the Republic of Mali. However.4 GHz frequencies without a license. The lesson is to do your research at the onset. Once you have identified your competitors. Be sure to collect anything that will help you determine how to position your network in the community. not just wireless. The competitive information you gather can include a list of services (including price and quality information). some countries restrict who can operate a network or require expensive licenses to do so. customer service techniques. First. Other countries.. etc. are more permissive. 2. This restriction makes it difficult for a small community to share a wireless network with other potentially interested parties or organizations. in such countries. their advertisements and marketing materials.286 Chapter 10: Economic Sustainability Research the Regulatory Environment for Wireless The regulatory environment for wireless networks also affects the type of business model that can be implemented. their target clients. telephone calls. Analyze the Competition The next phase in the evaluation of your community involves an analysis of the wireless network s competition. VoIP is prohibited for all networks.g.

you will lose $16. the subsidized center maintained low prices because it did not have to cover its costs.000 USD. and you will be able to use the computer for five years. office rental fees. Here is an example. This scenario eventually caused the locally owned center to go out of business. For example. the subsidized center went out of business as well. First. and you should set aside extra money for this purpose. In addition. you must determine the resources needed to start your project and the recurring operating costs. switches. to the costs to register your organization as a legal entity. If the initial cost to purchase the computer was $1. you can determine your competitive advantages in the community. These expenses can range from the initial investment you make in hardware. your annual depreciation will be $200 USD. telephone.Chapter 10: Economic Sustainability 287 It is important to evaluate your competition for many reasons. After the funding stopped. hubs. loans. including the cost of Internet access. Is there something that you can do better than the competitors to make your services more effectively fit the needs of the community? Finally. An advisable and very common method to deal with this is to take the price of the device and divide it by the period of time you estimate that it will last. There have been several instances where a subsidized telecenter was established by a donor organization in a small village with limited demand. a wireless network that can exclusively offer a faster Internet connection than a competitor is a competitive advantage that facilitates client utilization. Competitive advantages are those which cannot be easily replicated by the competition. To make your project sustainable. and equipment for access points. Knowing what already exists will allow you to determine how your network can contribute value to the community. Recurring costs are what you must pay to continue to operate your wireless network. and regular investments to replace malfunctioning or obsolete equipment. In other words. installations. etc. salaries. UPS. it is of fundamental importance that . cables. electricity. In one circumstance.67 USD every month so that you can eventually replace this computer. it helps you determine the level of market saturation. Every piece of equipment will eventually break down or become outdated at some point. despite the fact that there was already a locally owned cybercafé there. by analyzing your competitors from the customers point of view and understanding their strengths and weaknesses. equipment maintenance and repairs. This process is called depreciation. Start-up costs include everything you must purchase to start your wireless network. due to low revenues and high costs. An average computer is supposed to last for two to five years. Determine Initial and Recurring Costs and Pricing When you are planning to set up and operate your wireless network. analyzing the competition can stimulate innovative ideas for your service offerings.

integration etc. It is always better to over-budget for expenses than to under-budget. It is a good tool to structure the different costs. you should try to be very realistic about the life-cycle of all the implemented gear and plan for their depreciation carefully. and make realistic estimations on your recurring expenses. and it will help you to distinguish between initial costs and recurring costs. The following grid (continued on the next page) shows you a way to classify and list all of your costs. including yourself • Equipment maintenance and support costs for software. Categories of Costs Initial / start-up costs Recurring costs • Handling costs / salaries for employees or freelancer. hardware and ancillary equipment • Security personnel • Training costs (refreshers) • Check ups (analyses) and consultancies Labor costs • Development costs for programming. In any case.288 Chapter 10: Economic Sustainability you save the money to compensate for the depreciation of equipment each month. Some countries have tax laws that determine the period of depreciation for different types of devices. especially during the first year of operations as you learn how to better manage your network. With every wireless project. • Installation costs • Recruiting costs • Training costs (introduction) . there are always unforeseen costs. It is important to research all your start-up costs in advance. Try to find out all your costs in advance and make realistic estimations on your expenses. Keep these savings until you finally have to spend them for equipment replacement. testing.

g. capital costs for paying back your setup costs • Costs for advertising • Local fees • Legal and accounting services • Acquisition and production costs (for hardware like PCs. posters. it is generally best to maintain the lowest cost structure for your network.g. generator. data media. security grills) • Legal costs. etc. radio link equipment and software) • Ancillary equipment (e. stickers. Do not be afraid to think outside the box and be creative when determining the best solution. electrical wiring and boxes. Once again. .) • Rent or leasing rates • Depreciation of hardware and equipment • License fees • Consumables and office supplies (e. In other words. such as business registration • Initial license costs (VSAT) • Initial marketing costs (flyers. and shop around for the best deals on quality service. paper. etc. particularly the ISPs. lighting. binds. Depending upon demand for information access and ability to pay.. tiles and carpeting) • Premises costs (new building. keep your expenses as low as possible. UPS. Before installing an expensive VSAT. opening party) To improve your chances of sustainability. be certain that what you purchase from suppliers corresponds with the demand in the community. switches. cables and cabling.Chapter 10: Economic Sustainability 289 Initial / start-up costs Recurring costs • Operating costs for hardware and operating systems (Internet access. ensure there is a sufficient number of individuals and organizations in your community willing and able to pay for using it. clips) • Operational costs to maintain data protection and security • Insurance premiums • Costs for energy and to ensure power supply • Loan payments. Take time to thoroughly research all of your suppliers.. telephone. an alternative method of connectivity may be more appropriate. VSAT. air conditioning.) • Data protection and security Material (non-labor) costs • Start-up inventory (chairs. tables. modification. curtains.

11 802. it is highly important that you save enough to be able to do so. As mentioned before. the new “g” standard was created. Concurrently. you could be spending more on maintenance in the long run. . but rare until 2005 June 2008 (estimated) June 2009 (estimated) Typical Data Rate < 1 Mbps 5 Mbps 20 Mbps 23 Mbps 23 Mbps 75 Mbps Keep in mind the rapid advancement and changes in technology and think about how and when it may be time for you to reinvest in newer and cheaper (or better) devices to keep your infrastructure competitive and up-to-date. There is an interesting example of a major wireless internet service provider (WISP) who had more than 3. when necessary.11n Release Date 1997 1999 2003 1999. The larger and more complicated your infrastructure becomes. the WISP never managed to break even because it had to spend too much money to maintain all the access points. New competitors designed better and cheaper access points and offered faster Internet access for less money. Many times this relation is not linear but exponential. If you have a quality problem with your equipment once it is rolled out. The amount of money you will spend to maintain your ICT infrastructure is hard to guess. it can cost you an enormous amount of money to fix it. However.290 Chapter 10: Economic Sustainability Keeping your costs down should not be at the cost of quality.11y 802. As soon as the company had invested time and money to install the version of expensive first generation 802. Finally the first WISP was forced to close down the company. In addition. although it was initially the market leader. the company underestimated the short lifecycle of such devices.11b access points. your sales will decrease because the equipment is not up and running. the more financial and labor resources you must allocate for its maintenance.11b 802. Look at the following table to get a better picture on the fast development of wireless standards and equipment: Protocol 802.11a 802. Because lowquality equipment is more likely to malfunction.11g 802. ICT hardware tends to get cheaper and better as time goes on.000 access points in operation for a while.

receiving funding involves a competitive or a non-competitive process. Typically. The authors in this book are by no means trying to oversimplify this in depth system of rules . Secure the Financing Once you have determined your initial and recurring costs and created your financial plan. You need to list all of your initial and recurring costs and make some calculations to find out if your project can be sustainable. are also considered donors. These key tips will assist when making pricing decisions: • Calculate the prices you charge so that you cover all costs to provide the service. Most donors have complicated procedures surrounding the distribution of funding. and make sure your prices correspond with these It is absolutely essential to make a financial plan before you start. and the Canadian International Development Agency (CIDA). A donor is an organization that contributes funding and other types of donations to an organization or consortium of organizations to help them manage projects or support causes. The most traditional method of receiving funding for wireless networks in the developing world is through grants given by donors. Because this funding is provided in the form of grants or other donations.Chapter 10: Economic Sustainability 291 Once you have identified and mapped out your costs. Large foundations like the Gates Foundation and the Soros Foundation Network and private companies are other types of donors. The non-competitive process is more infrequent. including all recurring expenses • Examine the prices of your competitors • Evaluate what your customers are willing and able to pay for your services. The next step is to research and secure the appropriate amount of money to start up and run your wireless network. so this chapter will focus on the competitive process at a very high level. the United Kingdom s Department for International Development (DFID). it is not expected to be repaid by the organizations implementing the wireless projects or by the project s beneficiaries. Scientific and Cultural Organization (UNESCO). such as the United States Agency for International Development (USAID). Government agencies that specialize in international development. you know how much financing you will need to run a successful wireless network. Such donors include large international organizations like the United Nations (UN) and various specialized UN agencies like the United Nations Development Program (UNDP) and United Nations Educational. This is a complicated and time-consuming process to do correctly. you should also determine what and how to charge for your services.

292 Chapter 10: Economic Sustainability and regulations. Another way of accessing the necessary funds to start and maintain a wireless network is through microfinance. NGOs and other organizations compete through the submittal of their proposals. the donor creates a request for proposal (RFP) or a request for application (RFA). microcredit involves very small sums of money. sometimes a microcredit loan is not sufficient. They then rent the use of their mobile phones to community members on a per-call basis and earn enough money to repay their debt and make a profit for themselves and their families. savings and other basic financial services to the world s poorest people. which solicits various non-governmental organizations. Angel investors are normally wealthy individuals that provide capital for business start-up in exchange for a high rate of return on their investment. Sometimes donors also supply funding to support an organization s operations. Unfortunately. The use of microcredit to fund wireless networks does pose one constraint. During the competitive bid process. Pioneered in the 1970 s by organizations like ACCION International and Grameen Bank. The authors intend only to convey a general understanding of this process for communities attempting to establish wireless networks in the developing world. Usually. the process involves an individual or a group completing and submitting a loan application in the hopes of receiving a loan. there have been many other successful applications of microcredit that have brought technology and its value to the developing world. Many expect a board position and maybe a role in the organization. a type of microfinance. In response to this RFP or RFA. Because the ventures in which they invest are start ups and. and the lender. collateral or steady employment. private companies and their partners to submit proposals outlining their plans for projects within the constraints of the donors objectives and guidelines. An example includes the story of village phone operators. microcredit programs have been highly successful in many developing countries. Finally. Despite the fact that these individuals lack many of the traditional qualifications needed to obtain loans like verifiable credit. therefore. . which are then evaluated by the donors based on specific established criteria. or the provision of loans. because a large amount of capital is needed to purchase the initial equipment for wireless network set up. Another mechanism for getting funding to start a wireless network is angel funding. but this type of funding is more unusual than the competitive bid process. the individual or organization that provides the loan. enables poor individuals and entrepreneurs to receive loans in small amounts of money to start up small enterprises. These entrepreneurs use their microcredit loans to purchase mobile phones and phone credits. angel investors tend to expect different things in addition to their return. Typically. giving money on condition that it is returned with interest. the donor organization selects the most appropriate and highest ranking proposal to fund the project. However. often high risk. microcredit.

You can also note the areas where you can improve. sales. . marketing. Opportunities and threats are external. One tool often used to assist with this self-evaluation is an analysis of strengths. opportunities and threats.Chapter 10: Economic Sustainability 293 Some angels want to have a stake in the company. By understanding your strengths and weaknesses and comparing them to those of your competitors. Be sure to distinguish between where your organization is at the beginning of this endeavor from where it could be in the future. it is often challenging to find angel investors to help setup a wireless network. called SWOT. Afterwards. angels frequently ask the businesses not to make certain key decisions without their approval. identify local resources to fulfill these skills. First. Be sure to respond to the questions asked and list your strengths. in comparison to the competencies needed for a wireless project. including those of staff and volunteers. but not impossible. To conduct this analysis. you can determine your competitive advantages in the market. weaknesses. while others prefer shares in the company that can be easily redeemable at face value. opportunities and threats in the spaces designated. It is important to be realistic and honest about what you do well and what you are lacking. and operations. Because of the high risk involved in developing markets. which enable you to analyze real world conditions and how these conditions influence your network. human resources. To protect their investments. weaknesses. Your strengths and weaknesses allow you to evaluate your capacities internally and better understand what your organization can do. as well as its limits. accounting. and elaborate upon the external opportunities and threats in your community. specify your internal strengths and weaknesses. The diagram below will help you in creating your own SWOT analysis for your organization. make a list of all the competencies needed to run a wireless project successfully. The best way to find potential investors is through your social network and through research online. negotiation. and identify key gaps. Capacity areas should include technology. among others. Evaluate the Strengths and Weaknesses of the Internal Situation A network is only as good as the people who work and operate it. The team you put in place can mean the difference between success and failure. thus providing a clear exit for the investor. That is why it is important to reflect about your team s qualifications and skills. Map your team s skills sets to the competencies needed. legal.

you must often find innovative solutions to attain sustainability. It is essential to employ a model that capitalizes on opportunities and works within the constraints of the local environment. . you will better understand how to arrive at an appropriate model. you must refine your mission and service offerings. By exploring several examples and discussing the components of the models implemented in those instances.294 Chapter 10: Economic Sustainability Strengths • What do you do well? • What unique resources can you draw on? • What do others see as your strengths? • • • • Opportunities • What good opportunities are open to you? • What trends could you take advantage of? • How can you turn your strengths into opportunities? • • • • Weaknesses • What could you improve? • Where do you have fewer resources than others? • What are others likely to see as weaknesses? • • • • Threats • What trends could harm you? • What is your competition doing? • What threats do your weaknesses expose you to? • • • • • Putting it All Together Once you have gathered all of the information. To do this. All of the factors that you researched in the preceding steps come into play when determining your overall strategy. Based on the results of your external and internal analyses. you are ready to put everything together and decide upon the best model for the wireless network in your community.

However. founded by Baptist missionaries in 1904. economic. this Internet facility was also heavily used by the educated class in the community – the hospital's staff. this model provided a unique way of accommodating local community groups limited ability to pay for Internet services. The center required near total subsidization to operate and cover its costs. The largest expenses were electricity and Internet access. It is so remote that patients travel for weeks to get there often through a combination of travel by foot and by river. therefore. creative models needed to be constructed to reduce the telecenter s costs and provide access in a way that was sustainable. a project sponsored by USAID established a telecenter in this village to help improve education in this isolated community. and political issues that limited its sustainability. France and Canada. Various organizations in the com- . a traditional VSAT was used for connectivity. primarily technical. After reviewing the center s cost structure. it did have some shortcomings. has served as a hospital for many years. it is renowned for being an excellent facility and has had the support of German and American missionaries who have kept this facility in operation. This village. it was determined that it needed to cut its costs and look for new ways to increase its revenues. Although the center added great value to the community. Figure 10.Chapter 10: Economic Sustainability 295 In the distant jungles of the Democratic Republic of Congo. A study was commissioned to consider options for its future. and funding was to end by 2006. there is a rural hospital in a village called Vanga in the province of Bandundu. The center had been a great boon to the community. offering access to the world's knowledge and even providing consultation with distant colleagues in Switzerland. Although it is extremely remote. In 2004.1: Shared Internet over wireless In this instance.

Using pre-paid cards. This model has been deployed in villages in India. these communications are stored in the local kiosk s server. texts. this model addresses their communication needs in an innovative way. and voice mails. Additionally. Rwanda. When a bus or motorcycle with a mobile access point drives past a kiosk. In Vanga. Afterwards. including a hospital. they also share the costs associated with that connection. this model works well in Vanga because it has been tailored to meet community demand and leverage local economic resources.296 Chapter 10: Economic Sustainability munity share Internet access through a wireless network. including the billing and payment collection. and Paraguay. This model functions well due to specific conditions – namely an awareness and understanding of the value of the Internet among key community members. a community resource center. villagers are able to asynchronously send and receive emails. a pharmacy.2: DakNet's roaming access point Another example of a model adapted to fit the local context is that of First Mile Solutions DakNet. This arrangement enables the network of organizations to have a higher quality connection at a lower cost. have a need for Internet access and the means to pay for it. In the DakNet model. By taking into account the limited buying power of villagers. and a regulatory system that permits wireless sharing. several missionary groups. several organizations. technical maintenance and general business operations of the entire network. and local entrepreneurs are recruited and trained to operate kiosks equipped with Wi-Fi antennas. there is a franchise that exists in the country. and some non-profit organizations. Therefore. conduct web searches. and participate in e-commerce. the necessary resources to support Internet access. one organization in the village has the capacity and willingness to manage several aspects of the network s operations. Cambodia. the vehicle automatically receives . Figure 10.

with the appropriate skills to manage a business and willingness to accept certain requirements of the franchise organization. they need to have sufficient access to financial resources. relaying emails. Once the vehicle reaches a hub with Internet connectivity. etc. competition. reporting mechanisms. However. including an initial investment. working capital for certain recurring costs. to ensure this model will sustain itself. . This analysis should consider several key factors in the local environment. several key conditions need to be present. there should be sufficient demand for information and communication and few competitors in the community. management training. Conclusion No single business model will enable wireless networks to be sustainable in all environments of the developing world. messages. different models must be used and adapted as the circumstances dictate. Additionally. For such a model to be sustainable. it processes all requests. economic resources. and marketing tools. Finally. First. costs. you will help to ensure that your network brings value to the community in a way that corresponds with the users needs. Every community has unique characteristics. advice on start-up practices. Because these entrepreneurs are often asked to commit their own resources to the start-up costs. and sufficient analysis must be conducted at the onset of a project to determine the most appropriate model. by using the methods detailed in this chapter. this model requires a highly motivated and dynamic individual in the village. including community demand. DakNet integrates both mobile access and franchise models to bring value to people in remote villages. and shared files. a franchise organization must exist to provide financial and institutional support.Chapter 10: Economic Sustainability 297 the kiosk s stored data and delivers any incoming data. there are no guarantees of success. Although appropriate planning and execution will maximize the chances of making your network sustainable. standardized processes.

.

Specifically. the cost of locally produced materials and the cost of labour will be negligible. or Southern Asia. will be the easiest to replicate. This chapter describes some of our more memorable network projects. the right materials or equipment can be difficult and often im299 .11 Case Studies No matter how much planning goes into building a link or node location. 10th. This is the moment of truth that demonstrates just how accurate your estimates and predictions prove to be. you will still find that things do not always work out as you might have intended. Even after you install your 1st. or 100th node. whereas imported goods can be much more expensive when compared to its cost in the developed world. Whether you are about to embark on your first wireless project or you are an old hand at this. you will inevitably have to jump in and actually install something. Solutions that capitalize on local competitive advantages. it is reassuring to remember that there is always more to learn. and thus a process or solution designed for a more developed country may not be suitable in West Africa. Because transportation. but the price of an antenna might be double. It is a rare day when everything goes precisely as planned. Finding the right equipment is one of the most difficult tasks in developing markets. For example. communication and economic systems are not developed. namely cheap labour and locally found materials. General advice The economies of developing countries are very different from the developed world. one can manufacture and install a tower for a tenth of the cost of a tower in the United States.

300 Chapter 11: Case Studies possible to find. . Also. They also have the added benefit of being aerodynamic and uninteresting to passers-by. A fuse. comprised of a single 30mm diameter pipe which is then planted into cement. but they are made of poor materials and are thin. they can make very robust waterproof enclosures. it is often best to leave an exhaust hole at the bottom of the PVC enclosure. From old cars to televisions. and can save money. PVC tubing is far more resilient and is made to be waterproof. The typical mast is the 5 meter pole. This type of mast can be augmented by several meters with the use of guy lines. Equipment enclosures Cheap plastics are everywhere in the developing world. The author did find that leaving open holes can become a problem. It's best to construct the mast in two parts. the mast may be made with arms that can be securely cemented into a wall. or re-used. but requires the use of a ladder to complete and therefore some caution is suggested. thus mostly unsuitable for enclosing equipment. any material that has value will be stripped. The resulting space left around the equipment assures adequate air circulation. ownership. plant three lines 120 degrees apart. In West Africa. Antenna masts Recovering used materials has become an important industry for the poorest countries. A few quick adaptations and these same masts can be re-purposed for wireless networks. with a removable mast that fits into a base which is slightly larger in diameter. Using a wire mesh cover made from locally available screen material is advised to secure the exhaust hole from infestations. To sturdy the pole. and with end-caps that are torched-on. Alternately. you will see vehicles torn apart piece by piece and day by day. For example. Access points such as the Routerboard 500 and 200 can fit into such tubing. sized from 90mm to 220mm. This project is easy. In one instance ants decided to nest 25 meters above ground inside the PVC holding the access point. sold. Local metal workers will already be familiar with how to make television masts from scrap metal. thus finding wire that has a burn-up at a certain amperage and can substitute is a great advantage. is difficult to find. Finding local substitutes for materials also encourages local entrepreneurship. The resulting metal is sorted and then tossed into a truck to be sold. for example. forming an angle of at least 33 degrees with the tower. the most common PVC is found in plumbing.

Take your time and be selective in finding the right people for your project. Do not try to introduce a technology to a community without understanding which applications will truly will serve the community. In largely illiterate communities. Identify those people whom are likely to be opponents and proponents of your project. Have they been paying their bills? At times. and can speak the language of the community. alliances. Often. Often the community will have little idea how new technologies will help their problems. honest. Try to not build unwarranted enthusiasm. and not to make promises that you cannot keep. or community. take note of key players in an institution. champion and whom to diffuse. Regardless of the complexity of the undertaking. This is a difficult task and one that requires intimate knowledge of the institution or community. A "town-hall" meeting is often useful to see local politics. If you want to know the financial status of a company/organization. attempt to earn the support of the potential proponents and to diffuse the opponents. verify the facts that you are given. and other nontextual applications. In addition. but if the community is not involved the technology will not serve their needs. These people will already have the trust of the community. nor will it be accepted. Find the person. or persons whom are most likely to be interested in the project. Involving the community in a project can be the greatest challenge. or phone bill.Chapter 11: Case Studies 301 Above all: involve the local community Community involvement is imperative in assuring the success and sustainability of a project. ask to see an electricity bill. An effective strategy in gaining support is to find a respected champion whose motives are palatable. As early as possible. No other decision will affect your project more than having effective. potential beneficiaries will compromise their own values in hopes of winning funds or equipment. Thereafter. focus on digital to analog services such as Internet for radio stations. a community might be afraid and could subvert an initiative. frank. . It is important to be honest. Most often. or as members of a steering committee. and helpful. printing on-line articles and photos. it is easier to decide on whom to ally. a successful project needs support and buy-in from those it will serve. Moreover. Be careful in choosing your allies. local partners who trust you will be very frank. trusted local people on your team. If the project does not have a local ally. the project must take time to acquire this knowledge and trust from the community. Simply providing new features is useless without an understanding of how the community will benefit. and feuds in play. you will need to involve such champions as advisors. When gathering information. will know who to approach.

Ian Howard Case study: Crossing the divide with a simple bridge in Timbuktu Networks ultimately connect people together. and training cannot be gifts. There must be a negotiation. If a local ally and community buy-in cannot be had. where NGOs. This is where the low cost and mobility of a wireless network can be advantageous. Trying to do so can leave a project on unstable social ground. The project was deemed a success. which adds to the political challenges. another donor arrived with brand-new Pentium 4 computers running Windows XP. A project should ask for reciprocal involvement from the community. The students quickly abandoned the older computers and lined-up to use the new computers. the computers that are now sitting unused could have been deployed to another school where they would be used. law and policies are weak. Attempting to superimpose a network where human networks are not fully functioning is nearly impossible in the long term. The cost of Internet in less developed economies is high and the ability to pay is low. In many rural communities in under-developed economies. If they had been frank. training and funds. the project should consider choosing a different community or beneficiary. I once designed a project for a rural school in Mali. and contracts can be effectively meaningless. money. The community must be involved and they too must contribute. donors. other assurances must be found. the “no-go” option should always be evaluated. This is where pre-paid services are ideal. Savvy beneficiaries can earn handsome rewards by letting NGOs and donors lavish them with equipment. as they do not require a legal contract. Buy-in also requires that those involved invest in the project themselves. equipment. but shortly after the installation. Commitment is assured by the investment of funds before service is given. For example. and therefore always involve a political component. It would have been better to negotiate with the school in advance. and partners are not told of each others involvement with the beneficiary. It is important to know which other organizations are involved so you can understand how their activities might impact your own. Often. . My team installed an open source system with used computers and spent several days training people how to use it. threatening its existence. to know their commitment to the project. Above all.302 Chapter 11: Case Studies Another common pitfall is what I call "divorced parents" syndrome.

Timbuktu is widely known as an outpost in the most remote area of the world. This solution meant that the client site could get cheap Internet. Design Choices In this installation the client site is only 1 km away directly by line of sight. Being a symbol of remoteness. it added a small additional revenue source for the telecentre while not increasing its operational costs. Relative to other remote cities in the country. Though in an ideal situation it would be best to encourage more development of the small telecentre into an ISP. neither the telecentre nor the market were deemed ready. Timbuktu has a fair number of trained IT staff. at an affordable cost. Simple 14 dBi panel antennas (from http://hyperlinktech. This telecentre used an X. As is often the case. This model was chosen because it requires no significant investment by the supporting organization. the desert capital of Mali. and extends that network to a client site using a simple bridged network.Chapter 11: Case Studies 303 The author's team was asked by funders to determine how to connect a rural radio station with a very small (2 computer) telecentre to the Internet in Timbuktu. though having a world renowned name. it did not add a significant operational cost. flashed with OpenWRT and set to bridge mode. At last count there were 8 satellite connections into Timbuktu. which then relayed the connection back to Bamako. many projects have wanted to “stake a flag” in the sands of this desert city. a radio station. Two modified Linksys access points. were installed. the team decided to implement a model which has been called the parasitic wireless model. or adding complexity to the system. They currently use VSAT to link their telephone networks to the rest of the country. Thus. this solution minimized the initial investment while achieving two goals: first. While it added a source of revenue for the telecentre. three existing telecentres. At this site. Because of opposed usage patterns between an office and a telecentre there was no perceptible slowing of the network for either party. there are a number of information and communications technologies (ICT) activities in the area. The only configuration parameters required on both devices were the ssid and the channel. plus the newly installed telecentre at the radio station. and the other was installed 5 meters up the radio station's mast. precluding any private.com/) . The people Timbuktu is remote. Second. commercial interests from being sustainable. The city is to some degree over saturated with Internet. albeit not as fast or as reliable as a dedicated solution. Thus. it extended the Internet to the target beneficiary. One was installed on the wall of the telecentre. This model takes a wireless “feed” that is spliced from an existing network. SOTELMA and Ikatel. most of which service special interests except for the two carriers.25 connection to one of these telcos. there were serious concerns about whether this telecentre could become self-sustaining once its funders departed.

This model also allowed sharing of network resources. an existing antenna mast was used. The telecentre is located in the court of the Mayor's office. facing the client site. Having a bridge dedicated to one connection also simplified installation at the central site. This allows the author's team to simply ship a replacement and avoid the two day trek to Timbuktu. Though crude. An additional site will eventually be installed. . The bridge makes systems installed at the remote site (the radio station) appear as though they are simply connected to the local network. the telecentre simply unplugs the bridge on their side. At the client site. Training To support this network.304 Chapter 11: Case Studies were used. such as rebooting (power cycling) the access points. This was important because the radio station did not want to compete for clientele with the telecentre and the radio station's systems were primarily intended for the radio station staff. Though it is not optimal to bridge a network (rather than route network traffic). very little training was required. To disconnect the client. The telecentre staff were shown how to install the equipment and basic trouble shooting. clients can print at either site. meaning that this selective client base could support the cost of the Internet without competing with the telecentre. Because the client systems are on the same network. the access point and antenna were fastened using cement plugs and screws onto the side of the building. Financial model The financial model here is simple. The access point and antenna were mounted using pipe rings. about $30 per connected computer to the radio station. The telecentre charges a monthly fee. and it too will have its own bridge at the telecentre so that staff can physically disconnect the client if they have not paid. when technology knowledge is low and one wants to install a very simple system this can be a reasonable solution for small networks. this solution is effective and reduces risk that the staff would make a mistake while making changes to the configuration of the system. The telecentre also has the ability to easily disconnect the radio station should they not pay. while the telecentre has a color printer. so the principle client of the telecentre is the Mayor s staff. its supplier. This quick bridge reduced costs. At the Internet side. For example. This was many times cheaper than the alternative. the radio station has a new laser printer. and how to replace the unit should one fail. as the installation team was able to choose the best spot for connecting the client sites.

Local politics and competing subsidized initiatives are underway. and so the funder sponsored the installation of a VSAT system. and equipment and initial training was supplied as well. The project's goal was to provide information to the community in the hope that a better informed community would yield a healthier and more educated citizenry. with 8 computers. This rural city. a telephone and a digital camera. most of these services have just recently become available. A small two room building was built to house the telecentre. but this simple solution has proven to be an ideal use case. There are a number of VSAT systems now available to the region. Internet access is still an expensive undertaking in Timbuktu. providers are now . It has brought ICTs closer to a radio solution. The city slopes into the river gently. an all-in-one printer. a telecentre was installed in Gao. This connection was too slow and unreliable. which seems more more like a big village. but it seems the simplest solution provided the most benefit. The site received funding for all construction needed. but the cornucopic source of information for the centre is the Internet. It is located a bit outside of downtown. It is a standard telecentre. is Gao. As a result. Several months after its opening. It was meant to serve as a stop-gap measure while moving forward with a more complete solution. the telecentre was attracting few customers.Chapter 11: Case Studies 305 Summary The installation was considered an interim measure. Previously only C-band (which cover a larger area than Ku-band) systems were available. sits up the the river Niger just before it dips South crossing into Niger and onto Nigeria. Recently. fiber has been laid in almost every subway tunnel and canal throughout Europe. it has not yet led to building more physical infrastructure. It used a modem to dial-up to connect to an Internet provider in the capital. and reenforced local client/supplier relationships. and thus it has supplanted the more expensive satellite services. but the site was chosen because of its sympathetic host. films and radio. scanner. Ian Howard Case study: Finding solid ground in Gao One day's drive east from Timbuktu. and has few buildings taller than two stories. The centre provides information via CD-ROMs. in Eastern Mali. fax. In 2004. While it can be considered a success. It took the team several months of analysis and critical thought to arrive here. which is not an ideal location for attracting customers. The telecentre was expected to be self-sustaining after one year. As it stands.

The premise of this decision was to discourage a reliance on a short-term NGO. VSAT connections need to send packets up to the satellite and back down. Instead.306 Chapter 11: Case Studies redirecting their VSAT systems to new markets. which already had a 25 meter tower. A private contractor was hired. including middle and Western Africa. often introducing up to 3000 ms in delay for a round trip. perhaps twice as much. the connection provided 128 kbps down and 64 kbps up. who had been trained by the author to install a wireless system. A repeater was not desirable as it would introduce latency (due to the one-armed repeater problem) to an already slow connection. it was decided to use one radio to connect to clients. three approaches were discussed: installing an access point in repeater mode. The site was having trouble earning enough revenue to pay for this high monthly cost. a radio station. For simplicity it was decided to make that link a simple bridge. This collectively covered the costs of the VSAT. and the telecentre. but this investment has already begun to pay-off. This system would split the connection between three clients: a second beneficiary. and a second radio for to the dedicated backbone connection. we encouraged the telecentre to hire the local contractor to do it. so that the access point at the radio station would appear to be on the same physical LAN as the telecentre. each paying $140. In testing this approach functioned. The people Though capable and willing. This design proved to be fruitful. using the WDS protocol. After many different changes. This has led to a number of projects which use satellite systems for an Internet connection. To do this. avoiding the need to install towers at the client sites. We were able to reassure the client by agreeing to train and support the contractor in the fulfillment of this installation. as this tower was well above any obstacles in the city. and South Asia. This approach took much more time from the author's team. so the telecentre asked for help. After the VSAT was installed. or using a mesh routing protocol. and cost about $400 per month. and rather to build trust and relationships between domestic service providers and their clients. it was conceived that a backbone connection would be made to the radio station. and the extra revenue from the telecentre and the radio station would cover support and administration of the system. including replacing the access . To avoid this problem. its performance was dismal. though in the real world. the author s team did not do the actual installation. That tower would be used to relay to the other clients. Design choices Initially. Networks are still being installed and the author and his team are now home in Europe and North America.

These changes resolved both problems. To make the Internet more affordable. Thus. and so avoided interference from the FM transmission. the only feasible plan is to share the costs with other users. the team also realized the problem with the performance at the radio station: the radio frequency 90. After discussing many theories. Where there are few clients. using 802. For many IT for Development projects this expensive monthly cost is difficult to manage. removing the reliance on the radio station as a repeater likely made the installation more stable in the longer-term. shielded cable would be required. increasing the strength and reliability of the network. a rural community has only a few organizations or companies that could afford an Internet connection. Typically in Mali. This changed the frequency on the wire to 20 MHz. Often. and introduced massive packet loss. this is simply too expensive. foreign NGOs (Non Governmental Organizations). The team theorized that the towers on either side were too short. For this region. and to not use a relay site at the radio station. . Later. The installer then decided to place the access point at the telecentre directly using a small 3 meter mast. cutting off too much of the Fresnel zone. the FM signal (at 500 watts) was completely consuming the signal on the Ethernet cable. All sites were able to connect. the model developed by his team included anchor clients: clients whom are solid and are low-risk. Financial model The satellite system used at this site cost approximately $400 per month. Typically these projects can purchase equipment and pay for the establishment of a wireless network. It is necessary to find a model where the monthly costs for a network can be met by those who use. these connections became more erratic and even less stable. allowing a greater number of organizations to access the Internet while reducing the cost per client. the technician decided that there must be a software or hardware bug affecting this design. Eventually. the United Nations Agencies and large commercial enterprises are among the very few whom qualify. this site used wireless to share the Internet to the community. and the Internet connection cost is high.0 MHz was about the same as the frequency of the high-speed (100BT) Ethernet connection. and at the radio station the speed of the Ethernet was changed to 10 Mbps. either directly to the telecentre to the radio station. For most community telecenters or radio stations. The client sites were 2 to 5 km away. The client sites also required small masts in this design. or the frequency of the Ethernet link would need to be changed. though the connections were at times too feeble. The advantage of using mesh or WDS here would be that client sites could connect to either access point. While transmitting. during the dust season. The masts were then raised.Chapter 11: Case Studies 307 points.11b. but most are not able to pay for the cost of the network after a short period of time (including the recurring Internet costs and operational costs).

SMS services are the most reliable communication service because voice conversations tend to cut off in the middle and suffer heavy noise. Any non-routine work. Following the complete breakdown of the Nigerian railway system. such as adding a new client. Those clients could be disconnected and could resume their service once they can afford it again. . Lessons learned By sharing the connection. Kafanchan used to be known as a busy and thriving town as it was the host of one of the main junctions of the national railway. Therefore it was not imperative to teach the telecentre staff how to support the system in its entirety. Any revenue earned from the beneficiaries contributed to a windfall. for how long The contractor taught the telecentre technician the basics of supporting the network. Training needed: who. what. Today. which will allow subsequent ICT projects to build on this base. At the moment. the population of Kafanchan has been forced to go back to its original source of income. was contracted out. which was fairly rudimentary. When the railway industry was booming. three other sites have Internet access. no fixed telephony (PSTN) is available in the area and GSM only just arrived in 2005. and in addition. the GSM coverage is just as poor as the quality of the service. it is valuable to find the right local talent and to encourage them to build relationships with clients. which is agriculture.308 Chapter 11: Case Studies Among the clients selected for this project were three anchor clients. Kafanchan is a poorly connected area in terms of fixed telephony and Internet connectivity. and demand.000 people located 200 km northeast of Abuja. almost 80% of Kafanchan's populations relied on it in one way or another. A second beneficiary. the telecentre is now self-sustaining. A local implementor will be able to provide the follow-up support needed to maintain and expand a network. This activity is building local expertise. a community radio station. Ian Howard Case Study: Fantsuam Foundation's Community Wireless Network Kafanchan is a community of 83. or deposit for future costs. in central Nigeria. However. but was not counted upon due to the small margins that both of these community services operated on. who collectively paid the entire monthly cost of the satellite connection. Though it takes more time and perhaps more money. was also connected.

. how could anyone come up with the idea of establishing the first rural Wireless ISP in Nigeria there? Fantsuam Foundation did and they made it happen. health services. power backup systems and VoIP deployments. When NEPA is available on the grid. is more commonly known to Nigerians as "Never Expect Power Always". the population relies on expensive diesel generators or kerosene for illumination and cooking. has worked together with the Zittnet team to provide technical support for wireless communications. as well as the foremost rural knowledge economy driver in Nigeria. Only light bulbs can be fed straight to the grid power since they can handle the low voltage without damage. also know as Zittnet. is funded by IDRC. Project participants Given the challenging background of Kafanchan. non-governmental organization that has been working together with the community of Kafanchan since 1996 to fight poverty and disadvantage through integrated development programs.Chapter 11: Case Studies 309 Poor access to electricity brings further challenges to the people of Kafanchan. generally known as NEPA (National Electric Power Authority). it provides an unregulated voltage in the range of 100-120 V in a system designed for 240 V. ICT services and social development in rural communities of Nigeria. Kafanchan is receiving power from NEPA on an average of 3 hours per day. bandwidth management. faith-based institutions. small enterprises and individuals. Fantsuam Foundation is a local. Objectives The main objective of Zittnet is to improve access to communications in the area of Kafanchan by implementing a community wireless network. NEPA changed its name to Power Holding Company of Nigeria (PHCN). IT +46. The national electric power company of Nigeria. solar energy. the International Development Research Centre of Canada. Becoming the first rural wireless ISP in Nigeria was part of their mission to be a recognized leader in the provision of rural development initiatives. The Wireless ISP of Fantsuam Foundation. a Swedish based consultancy company focusing on ICTs for development. This voltage must be regulated to 240 V before most loads can be connected. In 2005. The network provides intranet and Internet access to local partners in the community. For the remaining 21 hours. The community network is formed by community-based organizations such as educational institutions. Fantsuam's focus lies on micro finance.

Simulations with the best existing solar data reveal that Kaduna State. the system is expected to produce not less than 6 KWh/ day. The network operation center (NOC) of the organization runs completely from solar energy.310 Chapter 11: Case Studies Power Backup System In order to provide a reliable service to the community. The NOC load has been separated from the rest of the load of Fantsuam to ensure a reliable power source to the critical infrastructure in the NOC. . which provides uninterrupted voltage stability. The system can charge from three different sources: a diesel generator. Figure 11. consisting of a deepcycle battery bank and 2 kW (peak) solar panels. The rest of the Fantsuam's premises runs from NEPA or the generator via the battery bank.1: 24 solar panels with a nominal power of 80 W have been mounted to the roof of the NOC to provide power to the system 24/7. A hybrid power system was designed for Fantsuam. Zittnet needed to be equipped with a stable power backup system that would make the network run independently of the NEPA. In the worst months of the year. where Kafanchan is located. Each of the solar panels (Suntech 80 W peak) provides a maximum current of 5 A (when the solar radiation is highest during the day). receives at least 4 sun peak hours during its worst months which stretch from June to August (the rainy season). and from NEPA when electricity is available. a solar array. even when the battery bank is running low on power.

6 V • Optimum operating voltage (VMP): 17.2 V • Short-circuit current (ISC): 5 A • Optimum operating current (IMP): 4. The total estimated power consumption of the NOC is 5.inveneo. .Chapter 11: Case Studies 311 The solar system has been designed to provide 12 and 24 V DC output in order to match the input voltage of all low power servers and workstations for NOC infrastructure and training classrooms. The solar panels used are Suntech STP080S-12/Bb-1 with the following specifications: • Open-circuit Voltage (VOC): 21.org/?q=Computingstation.65 A • Maximum power at STC (PMAX): 80 W (Peak) The minimum 6 KWh/day that feeds the NOC is used to power the following equipment: Device Access points Low power servers LCD screens Laptops Lamps VSAT modem Total Hours/Day 24 24 2 10 8 24 Units 3 4 4 2 4 1 Power (W) 15 10 20 75 15 60 Wh 1080 960 160 1500 480 1440 5620 The power consumption for servers and LCD screens is based on Inveneo's Low Power Computing Station. http://www.6 kWh/day which is less than the daily power generated from the solar panels in the worst month.

with good cooling capabilities for the batteries and the inverters. The NOC was designed to provide a place safe from dust. two solar regulators. The NOC uses natural methods and is made from locally available materials. The server space accommodates a rack unit for servers and a fan. The server room and battery room face south to improve natural cooling and to help keep the room at an appropriate temperature.2: The NOC is built by locally made laterite brick stones. a server room. The battery storage room hosts seventy 200 Ah deep cycle batteries. as well as five inverters (one of them pure sine wave). to avoid dust and overheating. The batteries are stacked vertically on a metal shelf structure for better cooling. The building is comprised of four rooms: a battery storage room.312 Chapter 11: Case Studies Figure 11. power stabilizers and DC and AC disconnects. a working space and a room for equipment storage. The room has no regular windows. Network Operating Center (NOC) A new Network Operating Center was established to host the power backup system and server room facilities. . produced and laid by youths in Kafanchan.

The NOC is unique for its kind in Kaduna State. The roof was designed with an inclination of 20 degrees to host the panels and limit corrosion and dust. In order to reach these clients. does not fear the height of the 45m tall tower as she is aligning the antennas hosted in the top of the tower. Physical infrastructure: A communication mast Most potential clients to Zittnet are located between 1 km and 10 km from the premises of Fantsuam. The NOC building is constructed of locally produced laterite mud bricks. To achieve this goal. The roof has also been strengthened in order to carry the extra load of 150-200 kg. natural cooling techniques have been introduced in the NOC design: small fans and extractors and thick walls of bricks (double width) in the direction of the sunset. Fantsuam established a communication mast on their premises.Chapter 11: Case Studies 313 The server room and the battery space require effective low cost/low energy cooling as they need to operate 24x7. Extra efforts have been made to keep the panels easily reachable for cleaning and maintenance. The material is cheap since it is frequently used and comes from the top layer of soil. In October 2006. The south side of the building hosts 24 solar panels in a shadow-free area on its metal roof. one of the staff of Zittnet.3: Omolayo Samuel. a 45m (150 foot) tall self- . Figure 11. The bricks are produced locally by hand using a low-tech pressing technique.

They come with an integrated multi-band sectoral antenna that can operate both in 2. In this way. The mast was equipped with grounding and lighting protection as well as a mandatory signal light. A lightning rod was mounted at the highest point of the mast to protect the equipment against lighting strikes. . The light is equipped with a photocell which enables automated switching based on the level of ambient light.8 GHz frequencies.314 Chapter 11: Case Studies standing mast was installed at Fantsuam Foundation.4: The network topology of Zittnet in October 2007. The Nexus PRO™ TOTAL series offers QoS for traffic prioritization and bandwidth management per client using the IEEE 802.4 GHz and 5. Wireless backbone infrastructure The wireless backbone infrastructure is built using SmartBridges multi-band access points and client units from the Nexus PRO™ TOTAL series.1-5. Figure 11. The units are designed for service providers and enterprises to establish high performance point-to-multipoint outdoor wireless links. A metal ring was buried at the base of the tower at a depth of 4 feet.11e compliant WMM (WiFi Multimedia) extensions. the light comes on at night and goes off during the day. The rod is made of pure copper and is connected to the earth ring at the base of the mast using copper tape. The signal light mounted at the top of the mast is a requirement from the Civil Aviation Authorities. All three legs of the mast were then connected to the grounding circuit.

no less than eight clients are connected to Zittnet. . Two months later. Clients that are located within the area between the dotted lines are connected to the sectoral antenna. The low buildings force us to mount the equipment at a fairly low height. Low buildings Most client premises are single-story buildings with a height of no more than 3 meters. Plans are underway to expand the wireless backbone by setting up two wireless repeaters. One repeater will be located in Kafanchan city using an existing NITEL tower to enhance the wireless coverage in the city center. This repeater will provide coverage to many surrounding towns and may even enable a long-distance link to Abuja. the topology of the network is a star topology with two access points in the communication mast at Fantsuam's premises. as clients can not afford to invest in small (10 m) masts to host the equipment.Chapter 11: Case Studies 315 Currently. Most installations make use of water tanks or a simple 3 meter metal pole attached to the wall of the premise. while the remaining clients are connected to the omnidirectional antenna. as physical access is not possible. Zittnet connected its first client in early August 2007. These clients include: • The general hospital • New Era Hospital • Jagindi Street Clinic (health clinic) • Zenith Bank (for private use) • Isaiah Balat (Internet café) • New World Hotel • Throne Room GuestHouse • Fulke Problems encountered A few problem areas that have been constantly present throughout the project are as follows. which is located about 7 km from Kafanchan. One access point hosts a 90 degree sectoral antenna (blue dotted lines) and the other access point provides omnidirectional coverage to the surroundings (red dotted rings). The second repeater will be established in the Kagoro Hills. a small mountain group with a relative altitude to Kafanchan of about 500m. Many houses have very weak roof structures which makes it impossible to mount equipment on the roof.

hotels or large shopping malls in western countries where end-users buy vouchers online and log in using an access code. This huge monthly cost of connectivity has been a big burden for Fantsuam and a constant stress of being unable to meet the monthly bill. As an alternative to the high risk "flat fee" model. this ends up being very expensive. This problem is almost always present in common accessories such as power sockets. At the moment. the shield of the UTP cable connecting the access point with the NOC will be grounded using grounding blocks and fasteners. a nearby lightning strike damaged equipment mounted on a mast. you often find yourself buying locally available merchandise that breaks even before it is put into operation. . The Zittnet team is currently working on improving the surge protection by adding extra coaxial surge arrestors. a lack of quality products on the market is a widespread problem across the whole African continent. RJ45 connectors. The system offers flexible Pay-As-You-Go charges over broadband VSAT Internet connections to countries across sub-Sahara Africa. Fantsuam had a subscription of 128/64 kbps dedicated bandwidth at a cost of $1800 USD/month. power bars. Further means need to be investigated to prevent damage to equipment caused by nearby lightning. CAT5 cabling. In September 2007. the market is flooded by "cheap" and very low quality articles. Since quality products are hard to find. Although the landscape in Kafanchan is very flat. This kind of access model is typically found in airports. Furthermore. As most sub-Sahara countries lack policies for quality assurance of imported goods. and other low-tech equipment. During 2006. Lightning strikes Heavy thunder storms are frequent during the rainy season in Kafanchan. as well as its power supply. Low Quality Equipment Unfortunately. Fantsuam has implemented a system called HookMeUP provided by Koochi Communications.316 Chapter 11: Case Studies When the equipment is mounted low. the access point and its PoE injector are grounded to the tower itself. Business Model The only alternative for Internet access in Kafanchan is via satellite. vegetation in the form of thick mango trees easily block the line-of-sight. As no sort of warranty exists for these minor purchases. the first Fresnel zone is not cleared and lower throughput is experienced.

08 .08 0.10 0. Whichever parameter runs out first will consume the voucher. at a cost of a higher price for the end user.00 USD / h USD / 700 MB 112. Reselling of vouchers (continuously) The voucher system is based on three parameters: access time.50 208. Installation of wireless hotspot at client's premise (one occasion per client) 5.87 1.10 91. Fantsuam is no longer stuck with a fixed monthly cost but has only to pay Koochi for the bandwidth they actually have consumed.00 89.80 1.95 1. Installation of client premises equipment (one occasion per client) 2. The Wireless Community Network now provides the Fantsuam Foundation with five sources of income: 1. Fantsuam buys vouchers from Koochi Communications and resells them to its local clients in Kafanchan. In this way.40 26.00 83.10 0.33 121. Leasing of wireless equipment (monthly cost per client) 3.00 416. The risk of buying expensive international bandwidth has now been transferred to the Internet provider instead of the end user. Fantsuam foundation now acts as a reseller of vouchers from Koochi and a supplier of wireless infrastructure to the end users. data limit and validity time.00 728.10 0.60 1.00 71. Access time 30 min 60 min 12 hours 24 hours 1 month 3 months 6 months 12 months Data limit (MB) 5 10 60 150 500 1600 3500 7500 Validity time 1 day 5 days 14 days 30 days 1 month 3 months 6 months 12 months Price (USD) 0.33 100.60 121.28 0.Chapter 11: Case Studies 317 The HookMeUP system offers a 512/256 kbps dedicated VSAT connection to Fantsuam (from their ground station in the UK).28 10. Reselling wireless equipment (one occasion per client) 4.20 67.

The client pays in advance (pre-paid model) with the result that Fantsuam will never end up in huge debt with the provider. it might be beneficial to go back to the flat-fee model again. are using the Internet access for professional and private use without reselling access to its clients. The pre-paid model is appreciated by many as it helps them to keep track of their expenditures. when the number of clients have increased and they can rely on a substantial monthly income from the wireless network. They will provide wireless Internet access to all of their rooms and offer access to Zittnet's uplink by reselling vouchers. Having a flat-fee model means that you are forced to sell a certain amount of bandwidth every month. Although there is room for improvement. One of the main limitations of the PAYG model is the lack of flexibility and transparency. Other clients. For example. However.318 Chapter 11: Case Studies The greatest advantage of this system is that Fantsuam Foundation no longer has the burden of a huge monthly bill for international bandwidth. His Internet café hosts 10 computers that all are connected to Zittnet. the advantage of avoiding debts is far greater than the disadvantages. Only when the user logs off will he/she be informed about how many minutes are left to spend. like the General Hospital and the Jagindi Street Clinic. The clients purchase vouchers from the owner with a margin of 25% over the price offered by Fantsuam. clients that do not have access to a computer connected to Zittnet can access the network though the PC's at Isaiah Balat's café. Fantsuam's income from reselling vouchers depends on how much bandwidth their clients consume. the business model seems to fit the local reality of Kafanchan and many other rural communities in Africa quite well. --Louise Berthilson . It is even used by electricity companies in some counties. The current PAYG system provides very little feedback to the user about consumed time or volume. Clients The clients are free to use the Internet access for any purpose. With the Pay-As-You-Go (PAYG) model. In return. The New World Hotel is another client that aims to create a similar business model but on a larger scale. The pre-paid model works well in Africa since people are familiar with this model from mobile operators. Isaiah Balat is reselling vouchers (that he bought from Fantsuam) to his clients. With time.

8 hours South-West by four-by-four from Bamako. The city is also endowed to be in one of the most fertile regions of the country.Chapter 11: Case Studies 319 Case study: The quest for affordable Internet in rural Mali For several years the international development community has promoted the idea of closing the digital divide. the Governor. To help mitigate some of the cost of bringing the Internet to rural areas of the developed world.11b wireless network that shares the telecenter's VSAT Internet connection with 5 other local services: the Mayor. In 24 hours the team installed an 802. Because the infrastructure does not exist. the district's Mayor's council (CC) and the community advisory service (CCC). the installation of a telecentre in the developing world seems like a realizable and worthwhile effort. Mali's main cash crop. Access to information and communications tools has been shown to have a dramatic impact on quality of life. Moreover. These clients had been selected during a reconnaissance two months prior. It was believed that this site would be the least difficult of the rural areas in Mali to make a self-sustaining telecentre. in its cotton belt. the author's team has promoted the use of wireless systems to share the cost of an Internet connection. This location is fortunate as hydroelectric power is much more stable and available than diesel generated power. the health service. this pilot was fraught with challenges. located on the margin of a man-made reservoir. Like many experiments. some rural communities are lucky to have any electricity at all. This rural city. an affiliated project asked the author's team to pilot such a wireless system at a recently installed telecentre in rural Mali. the capital. In November of 2004. While diesel generated power is far less stable. have electricity. The telecentre itself is housed at the community radio station. Radio stations tend to be great sites to host wireless networks in rural Mali as they are often well placed. They are also natural hubs for a vil- . few models have been shown to sustain these activities. Technologically it was a simple task. security and people who understand at least the basics of radio transmissions. holds water for the Manitali dam that powers a third of the country. During that visit the team had interviewed potential clients and determined which clients could be connected without complicated or expensive installations. This invisible chasm that has formed separating access to the wealth of information and communications technologies (ICT) between the developed and the developing world. For many donors fatigued by decades of supporting traditional development activities. this is much more expensive and difficult to do in the developing world than it is in the West.

And for a culture which is principally oral. and not so high as to interfere with coverage to client sites below in the bowl-like depression where most were found. The following calculation shows how to calculate the angle between the client site's antenna and the base station's antenna. access point.320 Chapter 11: Case Studies lage. A small switch was installed in the radio station. From the list of clients above. Design choices The technical team determined that the access point would be installed at 20 meters up the radio station tower. using standard trigonometry. 16 port switch). just below the FM radio dipoles. It was deemed unnecessary for this installation. The RB220 has two Ethernet ports. the network's champion. providing coverage to all client sites. across the yard. as there is considerable animosity and resentment between the various levels of government. assuring that the two clients less than a kilometer away could still receive a strong signal. the radio station itself has one Linux workstation installed by the author's project for audio editing. Some antennae have very narrow beam width and thus "overshoot" sites that are close. This proved to be a difficult mix. and the second that connects to the radio station's central . though at least two would be required and either a second radio or a channel splitter. From the main switch. The team then focused on how to connect each client site to this site.com/) would suffice.3m / 400m x = tan-1 (22m / 400m) x =~ 3 degrees In addition to the equipment in the telecentre (4 computers. one that connects to the VSAT through a cross-over cable. you will note that the clients were all government or para-governmental. tan(x) = + / difference in elevation height of base station antenna height of CPE antenna distance between the sites tan(x) = 5m + 20m . An 8 dBi omni (from Hyperlinktech. and there were continuing disputes regarding taxes and other fiscal matters. The 8 dBi antenna that was chosen has a 15 degree vertical beamwidth. Panel antennae were considered. Fortunately the director of the radio station. two cables run up to a Mikrotik RB220. Providing Internet to a radio station provides better information to its listeners. http://hyperlinktech. an Ethernet cable was run through plastic tubing buried at 5 cm across to the telecentre. was very dynamic and was able to wade through most of these politics. though not all. radio happens to be the most effect means to provide information. a laser printer.

users were able to quickly change a few settings. Users were encouraged to use secure protocols. You can find these parts at Mikrotik online at http://www. including: greater resilience to bad weather. lower contention rates (number of competing users on the same service) and it is more efficient at transferring data. Since the system ran Windows. including the Mikrotik software to control these networks. The RB 220 is housed in a D-I-Y PVC enclosure and an 8 dBi omni (Hyperlink Technologies) is mounted directly to the top of the PVC cap. a more thorough security system was left to be implemented in the future. The Mikrotik comes with a powerful command line and a relatively friendly and comprehensive graphical interface. The system has a powerful and friendly graphical interface that was superior to other products. firewall. These satellite systems are normally very reliable and are often used by ISPs. when time could be found to find an easier interface for controlling access.8. Atheros mini-pci a/b/g and POE was $461. designed for use as an access point or embedded computer. with License Level 5. in this case the dish was 2. and DNS-caching services.000 including installation. The system had no UPS or battery back up. The affiliate project had installed a C-band VSAT (DVB-S) system. A 128 kbps down and 64 kbps up Internet connection costs approximately $700 per month. providing DHCP. Mikrotik version 2.27. The AP has a virtual interface on each subnet and does all routing between. The installation of this VSAT was not ideal. Despite that the Mikrotik software requires licensing. imaps etc. Due to the above factors the team agreed to use these systems.2 meters in diameter and expensive.mikrotik. while routing traffic to the VSAT using NAT.com/routers. This system has several advantages compared to a Ku system though.php#linx1part0. thus. such as https. also allowing fire-walling at the IP layer. The network was designed to accommodate expansion by segregating the various sub-networks of each client.Chapter 11: Case Studies 321 switch. For this first phase. using a network sniffer like tcpdump one can see all traffic on the wireless link. It is a large unit. Note: this does not provide a firewall at the network layer. 24 bit private subnets were alloted. It controls the network. It is also expensive to operate. The RB220 runs a derivative of Linux. To limit access to subscribers. two PCMCIA slots. have two Ethernet ports. are temperature tolerant and support a variety of x86 operating systems. there was already a substantial user base in Mali. so once a power outage occurred the system would reboot and sit waiting for a . pops. There was little perceived security risk to the network. These access points are POE capable. costing approximately $12. the network uses MAC level access control. The total cost of the RB220. a mini-pci port. including adding a password to the default account. It is a small x86 based computer. a CF reader (which is used for its NVRAM).

The wireless system was fairly simple. proper configuration of the VSAT software as a service. We used a PowerNOC CPE bridge that cost $249. Each site had a part of the building that could physically see the radio station. Should it be required. as well as basic network troubleshooting. the CPE and the patch antenna were mounted on a small piece of wood that could be installed on the outside wall of the building facing the radio station. the team chose to use commercial. At the client site. At the client sites there were not local networks.322 Chapter 11: Case Studies password. Local staff were involved during the installation of the wireless network. the IP architecture could allow future partitioning and the CPE equipment supported STA mode. so the team also had to install cable and hubs to provide Internet for each computer. because the VSAT software was not configured as an automatic background service it did not automatically launch and establish the link. which had since been forgotten. Though the Cband systems are typically reliable. It lasted several weeks. Soon after the telecentre opened. the radio station wanted to display it. All of the client sites selected were within 2 km of the radio station. It was decided that because the client's networks were simple. hence it was not hidden from view. a POE made from a repurposed television signal amplifier (12V) was used to power the units. The team spent most of its time helping him learn how to support the basics of the system and the telecentre. that it would be easiest to bridge their networks. except for the cable installation. client grade CPEs: Based on price. The young graduate also required little training. They learned everything from wiring to antenna placement. An intensive training program followed the installation. which offered 20 hours of training and Internet use per month for only $40. To facilitate installation. this installation caused needless outages which could have been resolved with the use of a UPS. To make this situation worse. and was meant to teach the staff the day to day tasks. Wiring Ethernet networks is very similar to coaxial cable repairs and installations which the radio technician already performed regularly. In some cases the piece of wood was an angled block to optimize the position of the antenna.11b CPE bridge. Like all owners of new equipment. the Powernoc 802. and by limiting physical access to the modem. In some cases it was necessary to install Ethernet adapters and their drivers (this was not determined during the assessment). which the radio station technician quickly learned. Inside. small SuperPass 7 dBi patch antennas and home-made Power Over Ethernet (POE) adaptors. a bargain compared to the $2 an hour . students were lined up for the computer training. Preferably a space with glass doors would have kept the unit secure while keeping it visible. A young university graduate who had returned to the community was chosen to support the system.

The wireless system was included as a supplementary source of revenue because early financial projections for the telecentre indicated that they would fall short of paying for the VSAT connection. were not using the systems themselves. including the mayor. Lessons learned here were applied to other sites.Chapter 11: Case Studies 323 for Internet access. and the telecentre support tasks were divided among the staff. Bamako. The equipment was paid for by the grant. to help offset the costs to the community. Unfortunately. After much deliberation. despite their limited technical experience. Training did not stop there. and the only one who was trained in how to support the system. This was equivalent to one month of the service fee. Providing this training was a significant revenue and was a task that the young computer savvy graduate was well suited for. . and in consultation with the radio station that manages the telecentre. Most of the knowledge needed to operate the telecentre and network left with him. evaluating ways that this model could be improved. the young graduate left for the capital. The team realized the importance of assuring that the decision makers used the system. as described later. This did remove some of the mystique of the network and got the city's decision makers involved. Following training. and was mandated to be self-sustaining through the sale of its services. This left the telecentre effectively marooned. which served to assure their commitment. The radio station negotiated contracts with some support from its funding partner. Financial Model The community telecentre was already established as a non-profit. the principals. the team determined that it was best not to train another tech savvy youth. several clients were selected. but rather to focus on the permanent local staff. Our trainers have had to return for a total of 150 hours of training. and somewhat unsurprisingly. For this first phase. they too needed access. Based on the survey. had left. and provided training for them and their staff. Their most technically savvy member. Several people were taught each function. the program monitored the site and began to provide input. Deciding how much to charge was a major activity which required consultation and expertise that the community did not have in financial projections. clients were selected based on ease of installation and expressed ability to pay. Once the community services were connected. It seemed that although they were participating. after receiving an offer for a government job. This took much more time. Clients were asked to pay a subscription fee. but clients were still required to pay a subscription fee.

the first by the telecentre and the second for our internal analysis.324 Chapter 11: Case Studies To determine the monthly cost for an equal slice of bandwidth we started with the following formula: VSAT + salaries + expenses (electricity. This was just tolerable by the clients. This means that the community's income fluctuates wildly. the revenue foreseen in financial projections showed that even with an optimistic outlook. they would not only have trouble earning enough revenue on time to pay the . Unfortunately. Although they theoretically had the budget to pay for their service. Add another 10% to account for currency devaluation ($80). In trying to implement this model. In an agrarian society everything is seasonal. and so too is income. and looked feasible. $100 for salaries. $150 for electricity. they had long budget cycles with little flexibility. as they have limited bandwidth and only have room for those that can pay. the VSAT providers have little flexibility or patience. Because this was becoming complicated. we brought in business geeks. Thus. Cash flow management became a primary concern. to assure that the network can continue if a client defaults. and about $100 for supplies. who modified the formula as such: Monthly expenses + amortization + safety funds = total revenue The business experts were quick to point out the need of amortization of the equipment. the mayor signed on and used the backtaxes owed by the radio to pay for its subscription. at $100. supplies) = telecentre revenue + wireless client revenue We had estimated that the telecentre should earn about $200 to $300 per month in revenue. and had no room for complications. or if some equipment breaks. it was finally determined that amortization is a concept that was too difficult to convey to the community. About $750 in revenue from the wireless clients was required to balance this equation. amortized over 24 months) and the value of one client for default payments. and were broken down as: $700 for the VSAT. This amounted to roughly $150 from each client. As was soon discovered. Other fiscal complications arose as well. and that they would not consider that clients might default on payment. and that equals an expense of $1380 per month. Moreover.000. it would take many months for the payments to be made. This of course did not contribute to cash flow. First. or one could say "re-investment funds" as well as safety funds. For example. Total expenses were estimated to be $1050 per month. both formulae were used. This added about $150 per month for amortization (equipment valued at about $3. but required fair weather. as many public institutions were involved. regular payments are not part of the culture in rural Mali.

a major power surge caused by a lightning strike destroyed much of the equipment at the station. Ian Howard Case study: Commercial deployments in East Africa Describing commercial wireless deployments in Tanzania and Kenya. Following training the network operated for 8 months without significant technical problems. due to default payments. Conclusions Building a wireless network is relatively easy. due to the number of smugglers from Guinea and wayward rebels from the Ivory Coast. 99. A proper risk analysis would have concluded that a $700 (or even a $450) monthly payment left too narrow a margin between revenue and expenses to compensate for fiscal shortcomings. As projected. as well as concerns raised in this analysis. Unfortunately. Although cheaper. Before the project was able to find solutions to these problems. which by ignoring amortization and safety margins is affordable by the network. the network was not able to pay for the VSAT connection after the initial subsidized period. the telecentre was still off-line at the time that this book was written. due to technical problems. thereby suspending payment from their clients as well. it still sufficed for the size of the network. the cost of the VSAT already began to dig the telecentre into debt. Then. High demand and education needs limited the expansion of the network. including the access point and VSAT. the telecentre was not able to pay for its service and its service was suspended. the large C-band VSAT was replaced with a cheaper Ku band system. In this case. In contrast to projects devoted to ubiquitous access. we focused on delivering services to organizations. A payment model that considers re-investment and risk is a necessity. nor did it conform to social expectations. or eventually the network will fail. the payment model was not appropriate as it did not conform to fiscal cycles of the clients. After several months. but making it work is much more of a business problem than a technical problem. but getting the money to the Bamako-based bank also presented a problem. By that time this formula was finally deemed an unsuitable solution. typically those with critical international communications needs. I will . This system was only $450. As a result. Roads near the village can be dangerous. this chapter highlights technical solutions providing solid.5% availability Internet and data connectivity in developing countries.Chapter 11: Case Studies 325 fee.

transmission links. Connecting a handful of corporations to this WiLan proprietary 2. we negotiated with a local cellular company to place a PMP base station on their central mast. so we followed their lead.6 kbps SITA link (costing over $4000/month!). Frustrated by erratic PSTN services. and prevented broadcast traffic from suffocating the network. Commercial services. By late 2000. Market pressures kept prices down to about $100/month for 64 kbps. Bandwidth limits were configured directly on the customer radio. Tele2. Technical capabilities were developed in-house. had commenced operations with Breezecom (now Alvarion) equipment in 3. using fractional E1 transmission circuits for backhaul. and ERPs simply did not exist in East Africa. Our merger with the cellular carrier. we validated the market and our technical capacity to provide wireless services. Hungry applications such as peer-peer file sharing. and buoyed by a successful deployment of a 3-node point-multipoint (PMP) network for the Tanzania Harbours authority. As competitors haphazardly deployed 2. were 3-sector base stations installed.9 GHz. but a rising RF noise floor in 2. With grossly high PSTN international charges. contention ratios. Tanzania In 1995. towers were very tall. even though their wireless equipment purchase costs ranged from $2000-3000. etc. in mid-2000. but at that time (mid/late 2000) ISPs could operate with impressive. two facts emerged: a healthy market for wireless services existed. In most cases the small size of the cities connected justified the use of a single omnidirectional PMP base station. Leaf routers at each base station sent traffic to static IP addresses at client locations. limited to dialup email traffic carried over a 9.) so wireless data network design and deployment were straightforward.4 GHz networks. requiring staff training overseas in subjects such as SNMP and UNIX. one of the first ISPs in Africa. Dar es Salaam. I founded CyberTwiga. organizations rapidly shifted from fax to email traffic. Beyond enhancing the company . began in mid-1996. Infrastructure was in place (cellular towers. A sister company in the UK. summarizing key lessons learned over ten years in East Africa. and because the cellular partner operated an analog network.4 GHz system in late 1998.4 GHz would diminish network quality. clients were normally issued a single public IP address.326 Chapter 11: Case Studies describe two radically different commercial approaches to wireless data connectivity. very profitable. we had established coverage in several cities. included plans for a nationwide wireless network built on the existing cellular infrastructure (towers and transmission links) and proprietary RF spectrum allocations. voice.8/3. with Bill Sangiwa. Dar es Salaam is very flat. only in the commercial capital.

Customers were switched over to dialup in the month or so it took to reconfigure base stations and install new CPE. We quietly negotiated with the regulator. electrical mains supplies at base stations were less critical. To insure quality at customer sites. the regulator conducted a highly publicized seizure of our equipment. connectivity to 7 cities over 3000+km of transmission links. our fixed wireless data subscribers went offline. tightly tracking progress with job cards. and ultimately were rewarded with 2 x 42 MHz of proprietary spectrum in the 3. Infrastructure and regulatory obstacles often impeded operations.4/3. and later was determined to emanate from a Russian radar ship.Chapter 11: Case Studies 327 skills set. On the regulatory side. The cellular company tightly controlled towers. electrical power was always problematic. a top local radio and telecoms contractor executed installations. involved in tracking space activities. High temperatures. For the cellular company. and lightning were among the environmental insults tossed at outside plant components. Nairobi In early 2003 I was approached by a Kenyan company. We had to compete in a very limited IT labor market with international gold mining companies. AccessKenya. harsh equatorial sunlight. a major disruption occurred when the telecoms authority decided that our operation was responsible for disrupting C-band satellite operations for the entire country and ordered us to shut down our network. the cellular operator took the decision to close the Internet business in mid-2002. Despite backup generators and UPS systems at every site. Ultimately the network grew to about 100 nodes providing good. although not great. Only the merger with the cellular operator made this network feasible—the scale of the Internet/data business alone would not have justified building a data network of these dimensions and making the investments needed for proprietary frequencies. drenching rain. these training opportunities generated staff loyalty. Despite hard data demonstrating that we were not at fault. RF cabling integrity was vital. Cellular subscribers simply associated with a different base station. Of course the interference persisted.5 GHz bands. burdening our employees with the task of configuring many species of network hardware and topology. Customers often lacked competent IT staff. with strong UK business and technical backup to design and deploy a wireless . the UN. so that if there was a technical issue at a base station hours or days could pass before we gained access. Unfortunately. and other international agencies.

4/3. Also. Carrying traffic of multiple. high performance gear.5 GHz. and a single company could not hold both licenses. Since Nairobi is a hilly place with lots of tall trees and valleys. To meet the first regulatory requirement we chose to provide services using point-point VPN tunnels. and traffic transited our network in private IP space. Bandwidth was limited at the VPN tunnel level. Our criteria for wireless broadband equipment eliminated big pipes and feature-rich. At the customer site. With costs 100X greater than global fiber. “proprietary” frequencies. Also. 5. We judged that the bulk of our user population required capacity on the order of 128 to 256 kbps.5 GHz was expensive. Internet services were licensed separately from public data network operators. we selected Netscreen (now subsumed under Juniper Networks) as the vendor for VPN firewall routers. with a single base station. We started small. are carried by satellite.4/3.328 Chapter 11: Case Studies network in Nairobi and environs. about USD 120.8 GHz frequencies were subject only to an annual fee. as unique sales properties of our network. Broadband Access. Security and encryption added to network neutrality.000 year. improved wireless hardware. per deployed radio. The licensing overheads simply were not sensible. Form factor. and we were concerned about interference and the technical ability/political will of the regulator to enforce. were not exclusively licensed to a single provider.7/5. the network had to operate with total neutrality. Restated. All international Internet connections to Kenya in 2003. and flexibility. reliability. VirtualIT. and ease of installation and management were more important than throughput. competing ISPs or corporate users. We wanted demand to drive our . We selected Motorola s recently introduced Canopy platform in line with our business and network model. launching the “Blue” network. Two regulatory factors drove our network design. Ltd. Our network conducted a public-private IP conversion. and bigger market we designed a high availability network in line with regulatory constraints. At the time in Kenya. Benefiting from superb networking and business professionals. In contrast. wireless broadband networks demanded many base stations. a base station using 2 x 12 MHz attracted license fees of over $10.. satellite connectivity put a financial ceiling on the amount of bandwidth purchased by endusers. a private-public IP conversion delivered the globally routable address (or range) to the customer network. An ISP would deliver a public IP address to our network at their NOC. costing about USD1000 per MHz per year per base station. went live in July 2003. and at this writing. progress in internetworking. namely 3. spectrum in 3. Based on the operating experience of our sister UK company. not via a network of static IP routes.

Survey staff logged GPS positions for each client. softening initial capital expenditures. Rooftop deployments provided many advantages. Canopy has the advantage that the CPE and base station elements are light. We knew the tradeoff was that as the network expanded. Site surveys confirmed radio path availability and client requirements. unconstrained by night or rain. Technical staff became comfortable with customer support issues in a simple network environment. Big buildings also housed many big clients. In order to tightly control quality. permitted us to grow our network as traffic grew. We did not want to restrict landlords from offering roof space to competitors. rather than relying on a strategy of building big pipes and hoping we could fill them. Proper planning enabled completion of many installations in less than an hour. These became a key planning tool for base station placement. we categorically refused to . and it was possible to connect them directly into our core microwave network. rather to simply guarantee that our own services would not be interrupted. contractors under the supervision of a technical staffer performed installations. and third-party enhancements such as omnidirectional base stations. Cabling Canopy units was also simple. As a result all base stations were set up with two sets of cabling for all network elements. with outdoor UTP connecting radios directly to customer networks. Note that the point-point VPN tunnel architecture. the network was deployed on building rooftops.5% network availability. A typical PMP design. with base stations linked to a central facility via a Canopy high-speed microwave backbone. mains power and. so that most installations do not need extensive civil works or guying. As we compiled hundreds of customer GPS positions we began to work closely with a local survey company to overlay these sites on topographical maps. with its separate physical and logical layers. helped meet the goal of 99. Canopy. we would have to sectorize traffic and realign client radios.Chapter 11: Case Studies 329 network expansion. Rooftop sites did have the downside of more human traffic—workers maintaining equipment (a/c) or patching leaks would occasionally damage cabling. The gentle learning curve of a small network paid big dividends later. Technical staff attended two-day Motorola training sessions. protected the exclusivity of our radio frequencies. and contractor crews did not need any advanced training or tools. rather than have to deal with them on top of a complex RF and logical framework. not antenna towers. Following receipt of payment for hardware. and carried a laser rangefinder to determine height of obstacles. All leases stipulated 24x7 access for staff. a primary and a spare. Unlimited physical access. critically. required clients to purchase both wireless broadband and VPN hardware.

Graphed via MRTG or Cacti (as described in chapter six). and service was expanded to Mombasa. if their network was hit by a worm or if they didn t pay a bill) while the radio layer remained intact and manageable. but hugely justified as we operated the network. and gather valuable historical data.e. and verifying RF coverage enhanced our performance. . and flexibility. which will energize telecommunications when it happens. Lessons learned For the next few years satellite circuits will provide all international Internet connectivity in East Africa. Instead. RSSI. A benefit of the VPN tunnel approach was that we could prevent a client s traffic from passing over the logical network (i. Every client had the same hardware package. emphasis should be placed on reliability. It was not uncommon for clients to claim that service to their site had been interrupted for hours/days and demand a credit. After a number of customer issues (dropped VPN connections) were ascribed to power blackouts. Major investments in inverters and dual conversion UPS equipment at each base station were required to keep the network stable in the face of an erratic power grid. The Blue network combined a number of lessons from Tanzania with improved RF and networking technologies. Typical installations cost on the order of USD 2500. Compared to regions with fiber connectivity. Historical monitoring verified or invalidated these claims. Wireless broadband networks for delivery of Internet services therefore do not need to focus on throughput. potential deterioration of cable/connectors. parameters such as jitter. Tracing rogue operators. and traffic warned of rogue operators. Several groups have floated proposals for submarine fiber connectivity. bandwidth costs in East Africa will remain very high. and presence of worms in client networks. confirming the operating characteristics of equipment. As it grew from one base station to ten. Adding a portable spectrum analyzer to our initial capital investment was costly. but that compares to the $500-600 monthly charges for 64 to 128 kbps of bandwidth. the network RF design evolved and wherever possible network elements (routers) were configured with fallover or hot swap redundancy. redundancy. Fanatical attention to monitoring permitted us to uptweak network performance. we simply included a small UPS as part of the equipment package.330 Chapter 11: Case Studies permit clients to supply their own hardware—they had to buy from us in order to have service and hardware guarantees.

The point-to-point VPN tunnel architecture rolled out in Nairobi was extraordinarily flexible in service of client or network needs. Internet businesses generate a fraction of the revenue of mobile telephony. and so are marginal to the cellular companies. Radio failures were essentially unheard of. On the network side this translated into sizable investments in infrastructure substitution. with fail-over or hotswap capability is an expensive proposition in Africa.e building rooftops).Chapter 11: Case Studies 331 Reliability for our wireless networks was our key selling point. Implementing fully redundant networks. Trying to run a network on top of infrastructure that doesn t belong to you and is. probably 80% of our customer sites had two possible base station radios in sight so that if a base station were destroyed we could restore service rapidly. such as backup power. It was common for wellmanaged customer sites to remain connected for hundreds of days with zero unscheduled downtime. A key competitive advantage of our customer installation process is that we pushed contractors to adhere to tight specifications. from the point of view of the cellular provider. a goodwill gesture. On the RF side we had enough spectrum to plan for expansion. We could also sell multiple links to separate destinations. gives us the flexibility to replace the exist- . will make it impossible to meet service commitments. in theory. in our experience they raise more problems than they solve. Separating the logical and RF layers of the Blue network introduced an additional level of complexity and cost. and routinely tested. As attractive as potential alliances with cellular providers seem. In East Africa. configured for seamless fail-over. Consider the long-term reality that radio technologies will advance more rapidly than internetworking techniques. Nonetheless the core routers and VPN hardware at our central point of presence were fully redundant. With the growing number of base stations. as a single example. such as radios and power supplies. The most ordinary reasons for a single customer to lose connectivity were cabling or crimping issues. increasing the return on our network investments while opening up new services (such remote monitoring of CCTV cameras) to clients. Client connections could be set to burst during off-peak hours to enable offsite backup. Separating the networks. Similarly weekend staff members had access to an emergency cupboard containing spare customer premises equipment. Flexibility was engineered into both the logical and RF designs of the network. For base stations we took the decision not to install dual routers. but kept spare routers in stock. as well as cook up an alternative radio network design in case of interference. We controlled as much of our infrastructure as possible (i. We judged that the 2-3 hours of downtime in the worst case (failure at 1AM Sunday morning in the rain) would be acceptable to clients. and attention to details such as crimping and cabling.

cannot overcome the line-of-sight limitations presented by the mountains.: http://www. as conventional point-tomultipoint networks. By the end of February 2005. new mesh routers with multiple radio channel support are being developed and tested in Dharamsala. The total upstream Internet bandwidth available is 6 Mbps.D Case study: Dharamsala Community Wireless Mesh Network The Dharamsala Wireless-Mesh Community Network came to life in February 2005. The mesh backbone includes over 30 nodes. There are over 2. It is clear that scalability will become an issue if we continue to use a single radio channel. The broadband internet connection is putting the mesh under great load. More information • Broadband Access. while the “self healing” nature of mesh routing. Ltd.blue. with an emphasis on products that meet our technical requirements and our economically viable. Or we may install a different radio network in line with evolving technologies (Wimax) or client needs. mesh topology also offered much larger area coverage. Extensive testing during February of 2005 showed that the hard mountainous terrain is most suitable for mesh networking. Broadband Internet services are provided to all mesh members.virtualit. That is. while maintaining the logical network. what we got paid for. after all. following the deregulation of WiFi for outdoor use in India. the mesh had already connected 8 campuses.com/ • VirtualIT: http://www. the system seems to handle the load without any increase in latency or packet-loss. At present.: http://www.ke/ • AccessKenya.000 computers connected to the mesh. Finally.biz/ --Adam Messer. which is designed and built locally – known as the Himalayan-Mesh-Router .accesskenya.332 Chapter 11: Case Studies ing RF network without upsetting the logical network. all sharing a single radio channel. Ph. Ltd. The initial results are very promising. proved to be essential in places where electricity supply is very erratic at best. To solve this problem.co. The mesh network is based on recurring deployments of a hardware device. one must surrender to the obvious point that the exquisite networks we deployed would be utterly useless without unrelenting commitment to customer service.

The same mesh-routers are installed at every location. Allowing access to the mesh only by mesh-routers allows for strict access control policies and limitations to be enforced at the CPE (Client Premises Equipment) which is a crucial element needed to achieve end-toend security. and quality-of-service. The Asterisk PBX is also interfacing the PSTN telephone network.com/node/9).11b accesspoints at many of the same locations where mesh-routers are installed.11 dBi omnidirectional. Subscribers use a large variety of software-phones. Figure 11. to 12 . with only different antennas. Access to the mesh back-bone is only possible by mesh-routers.5: Dharamsala installer working on a tower The encrypted mesh back-bone does not allow access to roaming mobile devices (notebooks and PDAs). depending on the geographical locations and needs.24 dBi directional antennas and occasionally some highgain (and cost) sector antennas. due to legal issues it is presently used only for incoming calls into the mesh. software-based PBX is installed (Asterisk) and it provides advanced telephony services to members.airjaldi. traffic-shaping. However. The mesh channel is therefore encrypted (WPA). The mesh provides the backbone infrastructure while these APs provide access to mobile roaming devices. and also “hidden” to prevent mobile devices from finding it or attempting to access it. from 8 . . where needed. Simple wireless clients lack the intelligence needed to “speak” the mesh routing protocols and strict access policies. We use a wide range of antennas.Chapter 11: Case Studies 333 (http://drupal. so we have placed multiple 802. A central VoIP. as well as numerous ATAs (Analog Telephone Adaptors) and full-featured IP phones. The mesh is primarily used for: • Internet access • File-sharing applications • Off-site backups • Playback of high quality video from remote archives.

Many of the Dharamsala Mesh routers are powered solely by small solar panels. It is the capital of the state of Mérida. but local calls are charged by the minute and many villages lack phone lines altogether. on a plateau at about 1600 m. This makes them ideal for using with solar panels. The university operates a communication server with telephone lines to provide remote access to its network. The University of Los Andes (ULA) deployed the first academic computer network in 1989 which.334 Chapter 11: Case Studies Power consumption of the mesh-Router is less than 4 Watts.centuries-old university.6: Mérida is one of the three mountainous states of Venezuela. In 2006. as it very likely to survive when all other communication infrastructure is damaged.000 students. and home to a two. Figure 11. where the Andes reach 5000 m. has grown to encompass 26 km of fiber optic cable over which both a TDM and an ATM (asynchronous transfer mode) network are overlaid. a 50 km Gigabit Ethernet network has been deployed. --AirJaldi. many places in the city and the surrounding villages are out of reach of the fiber optic ring. The use of solar power in combination with small antennas and low power routers is ideally suitable for disaster areas. with some 35. . Nevertheless.com/ Case study: Networking Mérida State The city of Mérida lies at the foot of the highest mountain in Venezuela. http://airjaldi. despite economic limitations. over the same fiber optic cable.

Figure 11. While the rugged mountains of the region are a big obstacle for laying cables and building roads. performs modulation/demodulation and the assembly/disassembly of packets using a variation of the X. the cable car passes by an intermediate station called La Aguada.25 protocol known as AX. as well as several VHF (Very High Frequency) stations linked at 1200 bps that crisscrossed the country. The TNC is the interface between the analog radio and the digital signals handled by the PC. The TNC keys the Push To Talk circuits in the radio to change from transmit to receive.7: On its way to the peak. Packet radio Local amateurs operate a packet radio network. As early as 1987. a gateway would connect the local VHF packet network to stations overseas by means of HF stations that could span thousands of kilometers. called RedULA. Gateways between VHF and HF radios were built by attaching two modems to the same TNC and computer. which is 3450 m high and has an astounding view of the city of Mérida and other villages at distances up to 50 km. amateurs had a gateway with an HF (High Frequency) station working at 300 bps for contacts overseas. This task is aided by the existence of a cable car system. Typically. they can be helpful in deploying a radio network. Initially it worked at 1200 bps. which links the city to a 4765 m peak. reputedly the highest in the world.25. efforts to develop wireless access to the university's network. A national packet . albeit at a speed of only 300 bps. were undertaken from the very beginning.Chapter 11: Case Studies 335 For these reasons. The first attempts took advantage of the existing packet network operated by radio amateurs. using VHF amateur FM voice radios connected to a personal computer by means of a terminal node controller (TNC).

named after the call sign of its developer. and several radio amateurs were now able to access the Internet through the RedULA wired network. amateurs all over the world were soon able to connect to the Internet using different kinds of radios. This approach allows for much greater flexibility and easy upgrades. a government institution that promotes science and technology in the state. The need for two repeater stations underscores the limitations imposed by using higher frequency carriers. The limit on the radio bandwidth available on the VHF band puts a cap on the highest attainable speed. Amateurs are allowed to use 100 kHz wide channels using UHF (Ultra-High Frequency) signals. wrote the KA9Q program that implements TCP/IP over AX. One is located at La Aguada. KA9Q keeps the functions of the TNC to a bare minimum. a radio amateur with a strong background in computer networks. Phil Karn. Using this program. harnessing the power of the attached PC for most processing functions. which require line of sight to establish a reli- . one must move to higher frequency carriers. at 3600 m altitude.25. The digipeaters operated at 1200 bps and allowed for the sharing of programs and some text files among amateurs. we were soon able to upgrade our network to 9600 bps by use of more advanced modems. which relayed on digipeaters (digital repeaters. FUNDACITE also operates a pool of 56 kbps telephone modems to provide Internet access for institutions and individuals. at 2000 m. UHF antennas were built at LabCom. LabCom.336 Chapter 11: Case Studies radio network was also built.8: A UHF antenna for packet radio developed at ULA. A project was developed using this technology to link the House of Science in the city of El Vigia. to extend the network from Mérida to Caracas by means of just two such repeater stations. essentially a TNC connected to two radios with antennas pointing in different directions). Although El Vigia is only 100 km from Mérida by road. the mountainous terrain called for the use of two repeaters. and the other at Tusta. to Mérida and the Internet. In Mérida. the communications laboratory of ULA. Figure 11. The project was financed by FUNDACITE MERIDA.2 kbps modems doubled the transmission bandwidth. To increase that speed. Digital radios coupled with 19.

allowing for the simultaneous use of the same spectrum by several stations. While this was very exciting and certainly much cheaper than access through the telephone modems. without any radio. In the much lower VHF band. signals are easily reflected and can reach beyond hills. a faster medium would obviously be needed for a wireless backbone to connect remote villages. We therefore explored the use of 56 kbps modems developed by Dale Heatherington. • In DSSS the information to be transmitted is digitally multiplied by a higher frequency sequence. This meant the use of even higher carrier frequencies in the microwave range. but there is a hill in between that blocks radio signals. Fortunately. with the two antennas of the repeater pointing 40 degrees apart. but soon proved to be very useful in places where the electromagnetic spectrum is not overcrowded. These modems are housed in a PI2 card built by Ottawa amateurs. There are two ways to accomplish this: Direct Sequence Spread Spectrum (DSSS) and Frequency Hopping Spread Spectrum (FHSS). This scheme was tested to connect my residence to LabCom. it first found a use in civilian applications as a short-reach wireless local area network. at least on the backbone. . The distance is only 11 km. which is made by connecting two directional antennas back to back with a coaxial cable. Although this might seem to be a waste of bandwidth. and connected directly to a PC using Linux as the network operating system. which entailed high costs. While this system functions very well. the emergence of the World Wide Web with its plethora of images and other bandwidth-hogging files made it clear that if we were to satisfy the needs of schools and hospitals we had to deploy a higher bandwidth solution. Called spread spectrum. A connection was made by using a passive repeater to reflect off La Aguada. an alternative technology widely used in military applications was becoming available for civilian uses at affordable prices. allowing the bridging of distances of several kilometers. while at the same time allowing a number of users to share the medium by using different codes for each subscriber. Sometimes it is possible to reflect signals using a passive repeater. the recovery system is so efficient that it can decode very weak signals. thereby augmenting the transmission bandwidth. Spread spectrum Spread spectrum uses low power signals with its spectrum expanded on purpose to span all the allocated bandwidth.Chapter 11: Case Studies 337 able transmission.

2.338 Chapter 11: Case Studies • In FHSS.4 . held in Mérida in 1992. and the Technology Building at the University of Ile-Ife in Nigeria. including images and video clips.4835 GHz • 5. all sharing the nominal 2 Mbps bandwidth. the network was set up by ICTP staff with funding from the United Nations University and has been running satisfactorily ever since. The receiver must know this code in order to track the carrier frequency.4 GHz band to provide additional capacity.850 GHz In any of the above bands. Some subscriber sites that required longer cables between the antenna and the spread spectrum radio were accommodated by the use of bidirectional amplifiers. were soon exchanging files. The network started at LabCom. proving to be a much more cost-effective solution than the fiber optic network originally planned would have been. we were able to demonstrate this technique. We established some trial networks making use of external antennas built at the LabCom. the Venezuelan Ministry of Telecommunications opened up four bands for use with DSSS: • 400 . This ruling paved the way for the deployment of a DSSS network with a nominal bandwidth of 2 Mbps in the 900 MHz band. Later that year. This technology satisfied the needs caused by the surge in World Wide Web activity. Italy. This band can carry three simultaneously independent 2 Mbps streams. as the number of sites increased. During the First Latin American Networking School (EsLaRed '92). This provided a 90 degree beamwidth. but the effective range is lower than what can be achieved in the 900 MHz band. We started looking at the 2.960 MHz • 2. allowing transmission at several kilometers. We were very busy planning the extension of the backbone using .725 .5. maximum transmitter power was restricted to 1 Watt and the maximum antenna gain to 6 dBi. the transmitter is constantly changing its carrier frequency inside the allotted bandwidth according to a specified code. allowing many stations to share a certain portion of the spectrum. Physical Sciences Building. This group was aimed at providing connectivity between the Computer Center. Back in Mérida. Several subscriber sites. In 1993.512 MHz • 806 . LabCom housed an inhouse-built Yagi antenna pointed towards a corner reflector at Aguada. for a total EIRP (effective isotropic radiated power) of 36 dBm. These encouraging results were reported to a group set up at the International Centre for Theoretical Physics (ICTP) in Trieste. the observed throughput per user declined. in 1995. Both techniques exchange transmission power for bandwidth. illuminating most of the city of Mérida. where the connection to RedULA was available.

Broadband delivery system After visiting the Nashua.Chapter 11: Case Studies 339 2. . high density sectoral system. allowing applications such as streaming video at up to 30 frames per second. Each sector acts as an independent Ethernet LAN. facilities of Spike Technologies. Figure 11. At the base station. Frequency reuse on interleaved sectors makes for a spectrally efficient system. delivering a standard 10Base-T LAN connection to the subscriber. with a coverage of up to 50 km.9).9: Spike Technologies' full duplex. dramatically higher throughput. The narrowband digital radios can operate anywhere from 1 to 10 GHz. for the following reasons: Their broadband delivery system employs a special sectored antenna (Figure 11. The radios work with a variety of cable TV modems.10). Each sector transmits and receives on independent channels at 10 Mbps full duplex. the sectors are interconnected with a high-speed switch that has a very small latency (see Figure 11. with 20 dBi gain on each of up to 22 independent sectors.4 GHz when we found out about a start-up company that was offering a new solution that promised longer distances. we were convinced that their proprietary antenna and radio system was the best solution for the requirements of our state network. and the possibility of frequency reuse with narrowband microwaves. New Hampshire. for an aggregate throughput of 440 Mbps.

. with the base station located just above the cable car station of La Aguada at an altitude of 3600 m. At the subscriber site.10: Spike Technologies' system interconnections. Outside Equipment RF Cable Transceiver Intermediate Frequency Cable Modem PC 10baseT connection Figure 11. a similar radio and modem provide a 10BaseT connection to the local Ethernet.340 Chapter 11: Case Studies Cabling to transceivers Network Management Software High speed modem sector #1 High speed modem sector #2 High speed modem sector #3 High speed modem sector #4 High speed modem sector #5 Switch Internet • Conferencing • VOD • Broadcast Video Servers • Data • Mail • Web pages File Servers Telephony Servers Figure 11. a trial system was soon installed in Mérida.11: The subscriber end of the link. With funding from Fundacite.

with a beamwidth of 16 degrees each. 1998. In January 1999 we had 3 hospitals. a relief organization of the local government. Plans call for 400 sites to be connected within this year at full duplex 10 Mbps speed. and several governmental agencies. Figure 11. at 3600 meters. a videoconference between the penitentiary and the Justice Palace in Mérida proved that. In this case it was used for the arraignment of prisoners. where a satellite system provides Internet access. 2 newspapers. thus avoiding the inconveniences and risks of their transportation. 1 TV station. 6 educational institutions. educational system. The success of the trial prompted the state government to allocate the funding for a complete system to give high-speed Internet access to the state health system. 40 km from La Aguada.13 shows a map of the state of Mérida. aside from Internet access. . From La Trampa. 1 public library. and funding has already been allocated for this purpose. libraries. The fifth sector transmitted to a mountaintop repeater close to the village of La Trampa. Sector two served the Governor's Palace.12: Installation above Mérida at La Aguada. The first subscriber site was at Fundacite´s premises. Sector three served FUNDEM. and 20 social and governmental institutions sharing information and accessing the Internet.Chapter 11: Case Studies 341 Figure 11. On January 31. about 35 km from Mérida. Initially only 5 sectors were installed. while the light lines show the extension. community centers. 4 research institutions. Sector four served a penitentiary near the town of Lagunillas. The dark lines show the initial backbone. another 41 km link extended the network to the House of Science in the town of Tovar. the system could also support streaming video.

and sciences. • Health: The university hospital has a direct link to the intensive care unit. where a staff of specialist physicians is always on duty.342 Chapter 11: Case Studies Figure 11. . it is worthwhile to mention the following: • Educational: Schools have found an endless supply of material of the highest quality for pupils and teachers. languages. These doctors are available to be queried by their colleagues in remote villages to discuss specific cases.13: The Mérida State network Among the many activities supported by the network. Libraries have rooms with computers accessible to the general public with full Internet capabilities. and as a tool to communicate with other groups that share common interests. A group of researchers at the university is developing several telemedicine applications based on the network. Newspaper and TV stations have an amazing source of information to make available to their audience. especially in the areas of geography.

alleviating the overcrowding of the urban areas. and Latin America. who introduced me to Dr. allowing astronomers from all over the world access to the images collected there. • Government: Most government agencies are already connected and starting to put information online for the citizens. which was attended by Professor Jose Silva and Professor Luis Nunez from our university. This hands-on training lasted three weeks. I spent three months at Bellcore in Morristown. Given our very limited budget. the opportunities offered by the Net have a significant impact on the quality of their lives. EsLaRed'95 gathered again in Mérida with 110 participants and 20 instructors. We expect this to have a profound impact on the relationship of citizens with the government. attended by 45 participants from 8 countries in the region. Farmers have access to information about the commanding prices of their crops and supplies. Upon returning to Mérida.Chapter 11: Case Studies 343 • Research: The astronomic observatory of Llano del Hato. cited the Mérida broadband delivery network as winner of the SUPERQuest award in category 8-Remote Access as the best in that particular field of nominees. EsLaRed'97 had 120 participants. Maryland. management. Saul Hahn of the Organization of American States. We hope that this will help to reverse the trend of migrating out of the countryside. and it was endorsed by the Internet Society. located on a mountain at 3600 m and 8 degrees off the equator will soon be linked. who offered financial support for a training activity in Latin America. and wireless technologies were emphasized. SUPERCOMM '98. Training Since our earliest efforts to establish a computer network. and maintenance. we realized that training was of paramount importance for the people involved in the network construction. and three more months at the ICTP helping in the preparation of the Second Networking School in 1992. taking advantage of my sabbatical. Relief agencies and law enforcement agencies make heavy use of the network. These experiences allowed us to launch the First Latin American Networking School (EsLaRed'92) in Mérida. In 1990 the ICTP organized the First International School on computer network analysis and management. To this end. • Entertainment and Productivity: For people living outside the city. I spent the rest of my sabbatical at SURANET in College Park. Glenn Ricart. they proposed that we should somehow emulate this activity in our university. which also sponsored a Spanish and Portuguese first Networking . as well as improved agricultural practices. under the guidance of Dr. New Jersey. with instructors from Europe. the United States. Field researchers in many villages will enjoy Internet access. held in Atlanta in June. where I was joined by my colleague Professor Edmundo Vitale. we decided that we had to pool our resources with those of other people who also required training.

Phil." Caracas. Redondo Beach.eslared. • International Centre For Theoretical Physics. http://www. This is obviously exacerbated by the lower average income of people. It is hoped that this will contribute to the improvement of lifestyles in the fields of health." Sixth ARRL Computer Networking Conference.ictp. 29 August 1987. "Programme of Training and System Development on Networking and Radiocommunications.it/ • Escuela Latinoamericana de Redes. as well as create a more equitable relationship between citizens and government.trieste. and books. http://www. Italy." Trieste. "NORMAS PARA LA OPERACION DE SISTEMAS DE TELECOMUNICACIONES CON TECNOLOGIA DE BANDA ESPARCIDA (SPREAD SPECTRUM). faxes. Now ten years later. • Conatel. magazines. Concluding remarks The Internet has an even more profound impact in developing countries than elsewhere. References • Karn. • Heatherington.ve/ --Ermanno Pietrosemoli . owing to the high cost of international phone calls.org. education. held in Rio de Janeiro in 1998 with EsLaRed responsible for the training content. 1996. "A 56 kilobaud RF modem. Comision Nacional de Comunicaciones. CA. 17 November 1993.. "The KA9Q Internet (TCP/IP) Package: A Progress Report. Redondo Beach.344 Chapter 11: Case Studies Workshop for Latin America and the Caribbean. Ministerio de Transporte y Comunicaciones. 29 August 1987." Sixth ARRL Computer Networking Conference. EsLaRed continues to expand its training efforts throughout South America. Some dwellers in remote villages that do not have telephones are experiencing a transition from the 19th to the 21st century thanks to wireless networking. entertainment. and productivity. CA. D.

Orinoco. economic. Chilesincables.org Recent wireless data transmission technologies allow the creation of high speed. and Motorola. wireless technologies must be adapted and put to the best possible use. including IEEE 802. as well as permit free access to the knowledge gained. Such networks can bring great benefits to every user. and Minix.org endeavors to promote and organize wireless free networks in Chile in a professional way. Prism. we encourage utilization and exploration of a few vendors that allow for better control and adaptation to our needs (without necessarily increasing the prices). The hacker community has developed firmware that provides new functionality on this equipment. If these networks are built around the idea of removing restrictions to data access. Netgear. we employ Open Source operating systems. as well as some models of access points manufactured by Linksys. investigation. geographically separated networks at a relatively low cost. In order for free networks to flourish. OpenBSD.Chapter 11: Case Studies 345 Case study: Chilesincables. This kind of network is a direct response to the often restrictive commercial model ruling over much of our modern western society. FreeBSD. Even though a majority of wireless hardware available on the market will suit our goals. development and implementation of projects. they share our project s philosophy of being free technology with open source code. and Ralink. These include Wi-Fi cards with chipsets offered by Atheros. encouraging the adaptation of new technologies through adequate research. This is carried out by groups of hackers who do the research. etc. independent of the their political. including GNU/Linux. This fits our needs in the areas of routing as well as implementation of services such as proxies. We are also investigating recent innovations in the field. Description of technology We employ a variety of wireless technologies. the equipment has been modified in order to be accept external locally built antennas which meet local telecommunications regulations. In most cases. We do this by providing education about the related legal and technical aspects of wireless networking. For the network backbone itself. we call them free networks.11a/b/g. In addition. . web and FTP servers. and stimulating the adaptation of these technologies to meet the specific needs of Chilean communities and society. such as WiMAX. or social conditions.

etc. network s topology. each node must be associated to at least another node.346 Chapter 11: Case Studies Uses and applications The networks implemented so far allow the following tasks: • Transfer of data via FTP or web servers • VoIP services • Audio and video streaming • Instant messaging • Exploration and implementation of new services such as LDAP. This is generally provided by a captive portal. Administration and maintenance The operational unit of the network is the node. • Keeping the clients updated about the node s services (for example. a step that will allow the regulation of its internal administrative procedures and the formalization of the community in our society. new security methods. The users are free to use the net s infrastructure in order to create their own services. etc. by convention. or by technicians trained for this purpose. This allows the network to grow and to make more services available to every client. Chilesincables.org is currently in the process of acquiring legal organization status. how to get access to the network). selection of sites. name resolution. • Services provided by the clients. In addition. tasks related to deployment of new nodes. A node is maintained by an administrator who is a member of the community committed to the following tasks: • Maintenance of an adequate uptime (over 90%).) is carried out by the board of the community. . • Providing basic services (typically web access). Each node allows clients to associate to the network and obtain basic network services. The general administration of the network (specifically.

. Basic networking concepts are also taught. • The employment of materials and methods approved by the current regulations is a requirement for the normal development of the activities. all of our network administrators are trained in TCP/IP networking. training in radio communications techniques is essential. etc. To support these principles. Attendees are trained in the construction of antennas. knowledge of networking technology must be transferred to the users.org maintains a number of freeaccess documents and materials made available to people interested in a specific activity. free software gatherings. lectures. These include college workshops. • Promotion and Advertising.org undertakes the following activities: • Antenna Workshop. • Operating Systems Workshop.org considers training of its members and clients to be of vital importance for the following reasons: • The radio spectrum must be kept as clear as possible in order to guarantee adequate quality of wireless connections.Chapter 11: Case Studies 347 Training and capacity building Chilesincables. • Updating of Materials. • To ensure continuity in network operations. and introduced to basic concepts of radio communication. • In order to comply with Internet standards. Chilesincables. Chilesincables. The pictures on the following pages present a brief account of the activities in our community. Training on the implementation of routers and other devices based on GNU/Linux or other software such as m0n0wall or pfsense. Events for different communities that pursue our same goals are promoted. Therefore.

14: Omnidirectional slotted antenna workshop. .348 Chapter 11: Case Studies Figure 11.15: One of our staff members lecturing on the implementation of a m0nowall-based router in the administration of a node. In this session. Figure 11. attendants learned about building antennas and related theory.

Chapter 11: Case Studies 349 Figure 11.16: Detail of mini tower with samples of antennas. . cables and pigtails.

Figure 11.18: Location of the other end of the link.17: Wireless station and parabolic antenna used for the transmission of Santiago-2006 FLISOL via streaming video. .350 Chapter 11: Case Studies Figure 11.

The wireless transmission speed achieved was 36 Mbps at 1 km.19: Schematic representing Santiago-2006 FLISOL video streaming transmission.Chapter 11: Case Studies 351 Figure 11. using free software. .

This is one of the world’s highest nodes. and 30 clients.20: Quiani node. Its located at an elevation of 4000 m. . a Trevor Marshall 16+16 antenna. Figure 11. The node is connected to a downtown node more than 12 km away. about 2000 km north of the country’s capital.352 Chapter 11: Case Studies Figure 11.21: Node in southern Santiago. consisting of a 15 m tower.

22: Panoramic view of a node from the top of the tower. Figure 11. .23: Downtown node connected to the Santiago southern node.Chapter 11: Case Studies 353 Figure 11. Note the parabolic antenna for backhaul and the slotted antenna to connect the clients.

24: Implementation of node over a water tower in Batuco.25: Workshop on Yagi antennas organized by our community. . Metropolitan Region.354 Chapter 11: Case Studies Figure 11. Figure 11. Participants are building their own antennas. providing backhaul to Cabrati telecenter.

I finally located the town of El Baúl. Felipe Benavides (Colcad). David Paco (Dpaco). Venezuela already has some long range WLAN links. Although plenty of high grounds are to be found. It has an altitude of 4200 m and is about a two hour drive from my home town of Mérida.html). I found that there was no obstruction of the first Fresnel zone (spanning 280 km) between Pico del Aguila and El Baúl. in Cojedes State. My attention shifted to the Andes. I kept a record of the suitability of different spots for long distance communications. In the back of my head. like the 70 km long operated by Fundacite Mérida between Pico Espejo and Canagua. . Christian Vasquez (Crossfading). Eric Azua (Mr.cplus. in search of a stretch with high elevation at the ends and low ground in between. --Chilesincables. Daniel Ortiz (Zaterio). Using the free software Radio Mobile (available at http://www. Carlos Campano (Campano).11 Thanks to a favorable topography. Jose San Martin (Packet). in particular the famous "tepuys" (tall mesas with steep walls).org/rmw/english1.org Case study: Long Distance 802. Mario Wagenknecht (Kaneda). I first focused in the Guayana region. there were always obstacles in the middle ground. Andres Peralta (Cantenario). For several years. I have been traveling through sparsely populated areas due to my passion for mountain biking. Ariel Orellana (Ariel). Pico del Aguila is a very favorable place.Chapter 11: Case Studies 355 Credits Our community is made up of a group of committed volunteer associates among which are worthy of notice: Felipe Cortez (Pulpo). While looking at the terrain in Venezuela. it is necessary to find a path with an unobstructed line of sight and a clearance of at least 60% of the first Fresnel zone. Miguel Bizama (Picunche). Marcelo Jara (Alaska). Floppy). To test the limits of this technology. For the other end. Oscar Vasquez (Machine). Cesar Urquejo (Xeuron). whose steep slopes (rising abruptly from the plains) proved adequate to the task.

So. we would need 30 dBi antennas at both ends and even that would leave very little margin for other losses. the popular Linksys WRT54G wireless router runs Linux. non linearity is very severe and spurious signals are generated.26 using Google Earth: . due to the limited resolution of the digital elevation maps that are freely available. and can even be pushed up to 200 mW. Pico del Águila site survey On January 15. we felt confident that it could serve our purpose. but since the magnetic declination is 8° 16 . the Linksys can be located closer to the antenna than a laptop. So. In particular. which should be avoided. Sporting an output power of 15 dBm and receive threshold of -84 dBm. we could obtain a 5dB advantage compared with the Orinoco card. we decided to go with a pair of these boxes. Unfortunately. Another firmware. But at this value. Of course. On the other hand. One was configured as an AP (access point) and the other as a client. Therefore. We have been using Orinoco cards for a number of years. Furthermore. By setting the output power to 100 mW (20 dBm). when I looked towards 94°. DD-WRT. The Open Source community has written several firmware versions for it that allow for a complete customization of every transmission parameter. The WRT54G can operate at 100 mW output power with good linearity. we kept a spare set handy just in case. has a GUI interface and a very convenient site survey utility. The free space loss at 282 km is 149 dB. we looked at the equipment needed to achieve the goal. they are robust and trustworthy. 2006. after years of using it. I found the line of sight obstructed by an obstacle that had not been shown by the software.356 Chapter 11: Case Studies Action Plan Once satisfied with the existence of a suitable trajectory. The azimuth towards El Baúl is 86°. Several promising spots were identified. Although this is consumer grade equipment and quite inexpensive. This led me to refine my path selection. we settled for a pair of WRT54Gs. I rode my mountain bike for several hours examining the surrounding area looking for a clear path towards the East. as well as the output power. our antenna should be pointed to a magnetic bearing of 94°. resulting in the one depicted in Figure 11. and for each of them I took photos and recorded the coordinates with a GPS for later processing with the Radio Mobile software. OpenWRT firmware allows for the adjustment of the acknowledgment time of the MAC layer. I went to Pico Águila to check out the site that Radio Mobile had reported as suitable.

26: View of the 280 km link.27: Map and profile of the proposed path between Pico Aguila. and Morrocoy hill. Maracaibo’s Lake is to the West. and the Peninsula of Paraguaná is to the North.Chapter 11: Case Studies 357 Figure 11. The radio profile obtained with Radio Mobile is shown in Figure 11.27: Figure 11. near the town of El Baúl. .

so we tried an offset fed 2. Several tests were performed using various cantennas and a 12 dBi Yagi as a feed. The test site sits at 2000 m and therefore the elevation angle is 8°. as can be seen in the following picture: .4 m reflector. The importation costs are considerable.5° beam. This offered ample gain.4 GHz band are not available in Venezuela. In order to achieve a reasonable margin of some 12 dB for the link. We pointed the antenna at a base station of the university wireless network that was located 11 km away on a 3500 m mountain. so we decided instead to recycle parabolic reflectors (formerly used for satellite service) and replaced the feed with one designed for 2. we pointed the dish 14° downward. albeit with some difficulties in the aiming of the 3. The gain was way too low.5° offset meant that the dish appeared to be pointing downwards when it was horizontally aligned. The 22. Because of the offset feed.28: Propagation details of the 280 km link. We proved the concept with an 80 cm dish.28: Figure 11.4 GHz.358 Chapter 11: Case Studies The details of the wireless link are displayed in Figure 11. we needed antennas with at least 30 dBi gain at each end. Antennas High gain antennas for the 2.

For a meaningful measurement of the gain. Some of these include an annual training school on wireless networking (organized by ICTP) and another on computer . looking 14° down. who was immediately thrilled and eager to participate. The collaboration between the Latin American Networking School (EsLaRed) and the Abdus Salam International Centre for Theoretical Physics (ICTP) goes back to 1992. These instruments were also required for the field trip in order to align the antennas properly. In February 2006. I mentioned the project to my colleague Carlo Fonda. when the first Networking School was held in Mérida with ICTP support. Since then. but our efforts to measure the gain of the setup using Netstumbler were not successful. members of both institutions have collaborated in several activities. we looked for an antenna to be used at the other end.29: 2. We were able to establish a link with the base station at Aguada.4 m offset fed reflector with a 12 dBi antenna at its focus. we needed a signal generator and spectrum analyzer. While there. and also a pointing system better suited to the narrow radio beam. While waiting for the required equipment.Chapter 11: Case Studies 359 Figure 11. The actual elevation is 8° up. There was too much fluctuation on the received power values of live traffic. I traveled to Trieste to partake in the annual wireless training event that I have been attending since 1996.

This is shown in Figure 11.360 Chapter 11: Case Studies networks (organized by EsLaRed) that are hosted in several countries throughout Latin America.30 shows the disassembly of the mesh reflector. Figure 11. We also pinpointed the boresight for both the central fed and the offset antennas. Back at home. Sandro Radicella. we measured the maximum of the signal and located the focus. Accordingly.30: Carlo and Ermanno disassembling the satellite dish supplied by Mr. With a spectrum analyzer. Ismael Santos graciously lent his antenna for the experiment. and aimed the antenna at a signal generator that was located on top of a ladder some 30 m away. Mr. to support Carlo Fonda s trip in early April to Venezuela in order to participate in the experiment. We exchanged the feed for a 2. I found a 2.4 GHz one.31: . it was not difficult to persuade Dr. Ismael Santos.75 m parabolic central fed mesh antenna at a neighbors house. the head of the Aeronomy and Radio Propagation Laboratory at ICTP. Figure 11.

Carlo Fonda. It provides an unobstructed view towards El Aguila. The following day. . El Baúl Site Survey Once we were satisfied with the proper functioning and aim of both antennas. Of course. about 125 m above sea level.4 GHz feed We also compared the power of the received signal with the output of a commercial 24 dBi antenna. we found a hill (south of the town) with two telecom towers from two cell phone operators and one belonging to the mayor of El Baúl.31: Finding the focus of the antennas with the 2. The hill of Morrocoy is some 75 m above the surrounding area. given the weight of the antenna. but the value agreed with the calculation from the antenna dimension. there is some uncertainty about this value. There is a dirt road to the top. Gaya Fior and Ermanno Pietrosemoli reached the site on April 8th.Chapter 11: Case Studies 361 Figure 11. which led us to believe that the overall gain of our antenna was about 32 dBi. we decided to do a site survey at the other end of the El Baúl link. This showed a difference of 8 dB. a must for our purpose. We were receiving reflected signals.

given that the declination is 8° and therefore the true Azimuth is 268°. we asked Carlo to switch off the signal. At the same time. Once we were satisfied that we had attained the maximum received signal. Figure 11. Carlo removed the signal generator and replaced it with a Linksys WRT54G wireless router configured as an access point.M in El Baúl.32: Pico del Águila and surrounds map with Bronco truck. Javier Triviño and Ermanno Pietrosemoli traveled towards Baúl with the offset antenna loaded on top of a four-wheel drive truck. The Aguila team was able to install and point the mesh antenna before the fog and sleet began. To be sure we had found the proper source. After turning the signal generator on again. we installed the antenna and pointed it at a compass bearing of 276°. Lourdes Pietrosemoli and José Triviño) rode to the previously surveyed area at Pico del Aguila in a Bronco truck that carried the 2.33 shows the antenna and the rope used for aiming the 3° radio beam. Power for the signal generator was supplied from the truck by means of a 12 VDC to 120 VAC inverter.362 Chapter 11: Case Studies Performing the experiment On Wednesday April 12th. we were able to observe a -82 dBm signal at the agreed upon 2450 MHz frequency using the spectrum analyzer. Figure 11. the trace on the spectrum analyzer showed only noise. with assistance of Franco Bellarosa. Early the morning of April 13th. Indeed. At 11 A. Poor weather is common at altitudes of 4100 m above sea level. Javier substituted the spectrum analyzer on our end for another WRT54G configured as a client. This confirmed that we were really seeing the signal that originated some 280 km away.7 m mesh antenna. . the other team (composed by Carlo Fonda and Gaya Fior from ICTP. we performed a fine tuning in elevation and azimuth at both ends.

we started receiving "beacons" but ping packets did not get through. At once.33: Aiming the antenna at el Águila. Actual elevation was 1° upward. we began receiving packets with a delay of about 5 ms. It takes at least 2 ms for an acknowledgment to reach the transmitter. Fortunately.Chapter 11: Case Studies 363 Figure 11.34: El Baúl antenna installation. the OpenWRT firmware allows for adjusting the ACK timing.5°. This was expected. since the antenna has an offset of 22. Figure 11. since the propagation time of the radio wave over a 300 km link is 1 ms. . After Carlo adjusted for the 3 orders of magnitude increase in delay above what the standard Wi-Fi link expects.

36: Javier Triviño (right) and Ermanno Pietrosemoli beaming from the El Baúl antenna . Note the ping time of a few milliseconds.35: Screenshot of Javier’s laptop showing details of PDF file transfer from Carlo’s laptop 280 km away.35. Figure 11. using two WRT54G wireless routers. Figure 11. The results are shown in Figure 11. no amplifiers.364 Chapter 11: Case Studies We proceeded to transfer several PDF files between Carlo s and Javier s laptops.

17 April 2006. a team formed by Javier Triviño.Chapter 11: Case Studies 365 Figure 11. we found the time and resources to repeat it. This eliminates the need to wait for the 2 ms round trip propagation time on a 300 km path. . The latter is better suited for long distance point-to-point links since it does not require the reception of ACKs. One year after performing this experiment. José Torres and Francisco Torres installed one of the antennas at El Aguila site.. Alejandro González and Ermanno Pietrosemoli. 2007. The purpose of the modification of the standard WiFi MAC is to make it suitable for long distance applications by replacing the CSMA Media Access Control with TDMA. The other team. We used commercial 30 dBi antennas. Venezuela. This allowed for video transmission at a measured throughput of 65 kbps. On April 28th. installed the other antenna at El Baúl. formed by Leonardo González V. With the TDMA routers. the measured throughput was 3 Mbps in each direction. Eric Brewer of Berkeley University. This produced the total of 6 Mbps as predicted by the simulations done at Berkeley. and also a couple of wireless routers which had been modified by the TIER group led by Dr. A solid link was quickly established using the Linksys WRT54G routers..37: Carlo Fonda at the Aguila Site Mérida. Leonardo González G.

. but the link was not stable. when the received signal was about -78 dBm. However. the link was quickly established with the Linksys and the TIER supplied routers. in a place called Platillón. The proposed path is shown in Figure 11. Again. the measured throughput was a total of 6 Mbps bidirectional with the TIER routers implementing TDMA. with propagation times below 1 ms. with an average round trip time of 12 ms. We noticed considerable signal fluctuations that often interrupted the communication.38: Map and profile of the 380 km path. which pave the way for really inexpensive long distance broadband links. The TIER equipment showed no packet loss. the second team moved to another location previously identified at 382 km from El Aguila. This allowed for video transmission. The Linksys link showed approximately 1% packet loss.366 Chapter 11: Case Studies Can we do better? Thrilled by these results.38: Figure 11. Platillón is 1500 m above sea level and there is an unobstructed first Fresnel zone towards El Aguila (located at 4200 m above sea level).

It is particularly well suited for rural areas were the spectrum is still not crowded and interference is not a problem.Chapter 11: Case Studies 367 Figure 11. we are confident that Wi-Fi has a great potential for long distance broadband communication. . We'd also like to thank the Abdus Salam International Centre of Theoretical Physics for supporting Carlo Fonda s trip from Italy to Venezuela. Ismael Santos for lending the mesh antenna to be installed at El Aguila and to Eng. Acknowledgments We wish to express our gratitude to Mr.39: The team at el Aguila. provided there is good radio line of sight. José Torres (left). Javier Triviño (center) and Francisco Torres (right) Although further tests must be conducted to ascertain the limits for stable throughput. Andrés Pietrosemoli for supplying the special scaffolding joints used for the installation and transportation of the antennas.

Dr. and José Triviño. With the help of Franco Bellarosa. CPTM.368 Chapter 11: Case Studies Figure 11. Javier Triviño from EsLaRed. and Gaya Fior from ICTP. Dirección de Servicios ULA Universidad de los Andes and Fundacite Mérida contributed to this trial. RedULA.. This work was funded by ICA-IDRC. From left to right: Leonardo González V. Eric Brewer from Berkeley University provided the wireless routers with the modified MAC for long distance.. Leonardo González G. Ermanno Pietrosemoli and Alejandro González . Carlo Fonda. Sonesh Surana. as well as enthusiastic support through his collaborator. Lourdes Pietrosemoli. .40: The team at Platillon. The 2006 experiment was performed by Ermanno Pietrosemoli. For the 2007 experiments.

gov. http://wireless.eslared.ve/ • Abdus Salam International Centre for Theoretical Physics.funmrd.ictp.Chapter 11: Case Studies 369 References • Fundación Escuela Latinoamericana de Redes. http://openwrt. Latin American Networking School.ve/ --Ermanno Pietrosemoli .it/ • OpenWRT Open Source firmware for Linksys. http://www.org. http://www.org/ • Fundacite Mérida.

.

com/ • Pasadena Networks LLC.cacti.nec2. For more links and resources.superpass.freeantennas.com/ • Hyperlink Tech.net.com/ • Unofficial NEC2 code archives. http://www. http://www.sourceforge.free. http://www.nlanr.usbwifi.cookwareinc. http://www.com/EaKiu/ • EtherApe network traffic monitor.orcon.net/Projects/Iperf/ • iptraf network diagnostic tool.kiev. http://iptraf. http://netacad. http://dast.com/stest • EaKiu spectrum analysis tool. http://www.wlanparts.Appendix A: Resources We recommend these resources for learning more about the various aspects of wireless networking.cushcraft. http://hyperlinktech.ua/flowc/.com/nec/ • USB WiFi dish designs.nz/ Network troubleshooting tools • Bing throughput measurement tool.shtml • Cacti network monitoring package.htm • Free antenna designs.net/ • Flowc open source NetFlow collector. see our website at http://wndw.com/ • SuperPass. http://www. http://fgouget. http://www.dslreports.org/ 371 . Antennas and antenna design • Cushcraft technical papers on antenna design and radio propagation. http://www.seul. http://etherape.com/comm/support/technical-papers.net/.net/ • DSL Reports bandwidth speed tests. • Iperf network performance testing tool.nittany-scientific.org/ • Unofficial NEC2 radio modeling tool home page.fr/bing/index-en. http://www. http://www.

ee. http://www. http://www.ch/~oetiker/webtools/mrtg/ • My TraceRoute network diagnostic tool.debian.org/Network_Monitoring • Ntop network monitoring tool. http://openvpn.ee.linuxjournal.ee. http://en.html . http://www.ch/~oetiker/webtools/smokeping/ • SoftPerfect network analysis tools.org/ • NetFlow. the Cisco protocol for collecting IP traffic information.com/ • Driftnet network monitoring utility.com/~chris/driftnet/ • Etherpeg network monitoring utility. http://tldp. http://www.html • OpenSSH secure shell and tunneling tool.372 Appendix A: Resources • MRTG network monitoring and graphing tool.net/ • Network monitoring implementation guides and tutorials.nagios. http://openssh. http://people. http://www.softperfect.net/howto. http://www.org/ • RRDtool round robin database graphing utility.ethz.mil/ftp/pub/ttcp/ • Wireshark network protocol analyzer. http://www.com/article/7949 • Lavasoft Ad-Aware spyware removal tool.org/apps/all/Networking/Security_/_Admin.de/ • Linux security and admin software. http://www. http://people.org/ • Introduction to OpenVPN. http://www.com/ • Squid transparent http proxy HOWTO.org/ • OpenVPN encrypted tunnel setup guide. http://www. http://ftp.spychecker.com/ • Anti-spyware tools.org/wiki/Netflow • ngrep network security utility for finding patterns in data flows.wireshark. http://people.nl/mtr/ • Nagios network monitoring and event notification tool.arl.org/ Security • AntiProxy http proxy circumvention tools and information.html • ttcp network performance testing tool. http://wiki. http://ngrep.etherpeg.ch/~oetiker/webtools/rrdtool/ • SmokePing network latency and packet loss monitor.sourceforge.ethz.org/HOWTO/TransparentProxy. http://www.lavasoft.ethz.antiproxy.ntop.ex-parrot.linux.bitwizard. http://www.wikipedia.

com/papers/others/rc4_ksaproc.sourceforge.11 Wireless Network has No Clothes. http://winscp.html • Pittsburgh Supercomputing Center s guide to Enabling High Performance Data Transfers. http://lartc.org/ • Weaknesses in the Key Scheduling Algorithm of RC4. http://www.nl/ • Sawmill log analyzer. http://www. http://www.org/inet97/ans97/cloet.net/ • Security of the WEP algorithm.bristol. http://www.net/ • Your 802.info/pubs/bandwidth/index.cs. http://www.putty.edu/~waa/wireless.microsoft.com/ 373 Bandwidth optimization • Cache heirarchies with Squid.privoxy.ac.stunnel.net/latest/html/c2075.html • dnsmasq caching DNS and DHCP server. http://www.inasp.htm • Fluff file distribution utility.ps • Windows SCP client.org/ • Optimising Internet Bandwidth in Developing Country Higher Education. http://www.org. http://www.uk/fluff/ • Linux Advanced Routing and Traffic Control HOWTO.isaac.crypto.isoc. http://www.isaserver.berkeley. http://www.html • Enhancing International World Wide Web Access in Mozambique Through the Use of Mirroring and Caching Proxies. http://www.org/ • Microsoft Internet Security and Acceleration Server.psc.torproject.zonelabs.html • Stunnel Universal SSL Wrapper.cs.pdf • ZoneAlarm personal firewall for Windows.edu/isaac/wep-faq. http://www. http://squid-docs. http://www.Appendix A: Resources • Privoxy filtering web proxy. http://www. http://planetmy.html • Planet Malaysia blog on bandwidth management.sawmill.org/ • PuTTY SSH client for Windows.thekelleys. http://www.org/ • TOR onion router. http://www.edu/networking/perf_tune.umd.com/isaserver/ • Microsoft ISA Server Firewall and Cache resource site.uk/dnsmasq/doc.com/blog/?p=148 .

org/ • Real-time OLSR topology viewer.5 chipset.olsr. http://meshcube. http://www.pl • AirJialdi Mesh Router. http://squid-cache. http://www.macstumbler. http://kismac. http://www. http://pyramid.metrix.org/rfc/rfc3135 • Squid web proxy cache.org/nylon/utils/olsr-topology-view.edu/roofnet/doku. http://m0n0. http://cuwireless. http://www. http://madwifi.ch/wall/ • MadWiFi wireless driver for the Atheros chipset.freifunk.dd-wrt.org/ • Metrix Pyramid wireless router OS.polarcloud.kismetwireless.epitest.net/n9zia/wireless/page09.html • KisMAC wireless monitor for Mac OS X.org/ • Tomato wireless router OS for Linksys access points.com/node/9 Wireless operating systems and drivers • DD-WRT wireless router OS.net/wiki/FreifunkFirmware • MIT Roofnet Project. http://www.airjaldi.net/ • MacStumbler wireless network detection tool for Mac OS X.com/ .net/download • Freifunk OLSR mesh firmware for the Linksys WRT54G.php • OLSR mesh networking daemon.com/tomato Wireless tools • Chillispot captive portal.org/ Mesh networking • Champaign-Urbana Community Wireless Network software.ietf.qsl. http://www.com/ • HostAP wireless driver for the Prism 2.macpirate.374 Appendix A: Resources • RFC 3135: Performance Enhancing Proxies Intended to Mitigate LinkRelated Degradations.net/ • OpenWRT wireless router OS for Linksys access points. http://www. http://www.ch/ • Kismet wireless network monitoring tool. http://openwrt.csail.chillispot. http://pdos. http://drupal. http://www.info/ • Interactive Wireless Network Design Analysis Utilities. http://hostap.mit.fi/ • m0n0wall wireless router OS.

http://sourceforge.Appendix A: Resources 375 • NetStumbler wireless network detection tool for Windows and Pocket PC.w1ghz. http://www.org/rmw/ • Terabeam wireless link calculation tools. http://nocat. http://www. http://www.com/ • Homebrew wireless hardware designs.blue.com/ • NoCatSplash captive portal.net/index.net/projects/phpmyprepaid/ • RadioMobile radio performance modeling tool. http://nocat.net/download/NoCatSplash/ • PHPMyPrePaid prepaid ticketing system. http://www. http://seattlewireless. http://seattlewireless.net/HardwareComparison • Stephen Foskett's Power Over Ethernet (PoE) Calculator.net/ • WiFiDog captive portal.terabeam.gweep.com/ • SeattleWireless community wireless group.org/ • Wireless Network Link Analysis tool by GBPRR.cgi General wireless related information • DefCon long distance WiFi shootout.dk consultancy and services.cgi/LinksysWrt54g • NoCat community wireless group.net/ • Ronja optical data link hardware.wifidog.athenet.virtualit.com/support/calculations/index. http://www.org/ • Linksys WRT54G resource guide. http://www.cplus. wireless broadband carrier. http://ronja.net/~sfoskett/tech/poecalc.net/~multiplx/cgi-bin/wireless.dk/ .less.netstumbler. http://my. http://www. http://www.html Networking services • Access Kenya ISP.twibright. http://www.org/ • Linksys wireless access point information. http://www. http://linksysinfo.php • Wellenreiter wireless network detection tool for Linux.less.seattlewireless.main.wellenreiter.accesskenya.wifi-shootout. http://www.ke/ • Virtual IT outsourcing.com/ • Broadband Access Ltd. http://wire.net/ • SeattleWireless Hardware comparison page.biz/ • wire. http://www.co.

ac.org/ • RFC 1918: Address Allocation for Private Internets. http://www.org/ • NodeDB war driving map database.nodedb.netfilter. Uganda.ubuntu.ictp.376 Appendix A: Resources Training and education • Association for Progressive Communications wireless connectivity projects.trieste.it46.wifimaps.metageek.org/ Miscellaneous links • Cygwin Linux-like environment for Windows.inasp.partimage. http://www.org/ • ICTP bandwidth simulator.it/hherold/ • WiFiMaps war driving map database.ug/ • Radio Communications Unit of the Abdus Salam International Center for Theoretical Physics.cygwin.trieste.org/documentation/HOWTO/ networking-concepts-HOWTO.net/ . http://www.apc. http://www. http://wireless.com/ • WiSpy spectrum analysis tool.virgilio.org/wireless/ • International Network for the Availability of Scientific Publications. http://www.ordb.com/ • VoIP-4D Primer. http://xoomer. http://www.com/ • Graphvis graph visualization tool.php • wget web utility for Windows. http://www. http://www. http://www.ictp. http://www.wsfii.imagemagick.it/ • World Summits on Free Information Infrastructures. http://www. http://www. http://www.org/ • Partition Image disk utility for Linux.com/ • Open Relay DataBase.graphviz.it/simulator/ • ImageMagick image manipulation tools and libraries.ietf. http://www. http://www.makerere.org/rfc/rfc1918 • Rusty Russell's Linux Networking Concepts.html • Ubuntu Linux.se/voip4d/voip4d. http://wireless. http://www.info/ • Makere University.

Jack Unger. W. Addison-Wesley. ISBN #0-596-10144-9 .11 Wireless Network Site Surveying and Installation. Richard Stevens. Volume 1. ISBN #0-87259-904-3 • The ARRL UHF/Microwave Experimenter’s Manual. Rob Flickenger and Roger Weeks. 2nd Edition. O Reilly Media. Matthew Gast.net/. Cisco Press.Appendix A: Resources 377 Books • 802. ISBN #0-201-63346-9 • Wireless Hacks. http://bwmo. 2nd Edition. Rob Flickenger. ISBN #978-0-9778093-1-8 • Deploying License-Free Wireless Wide-Area Networks. ISBN #0-87259-312-6 • Building Wireless Community Networks. ISBN #0-596-10052-3 • 802. Bruce Alexander. O Reilly Media. Dean Straw (Editor).11 Networks: The Definitive Guide. Cisco Press. 2nd Edition. American Radio Relay League. American Radio Relay League. A free book about bandwidth optimization. 20th Edition. R. O Reilly Media. ISBN #0-596-00502-4 • How To Accelerate Your Internet. ISBN #1-587-05069-2 • TCP/IP Illustrated. ISBN #1-587-05164-8 • The ARRL Antenna Book.

.

11a and 802.447 2.427 2. Note that while all of these frequencies are in the unlicensed ISM and U-NII bands. so always check your local regulations before transmitting.Appendix B: Channel Allocations The following tables list the channel numbers and center frequencies used for 802.11b/g. and 20MHz wide in 802.462 2. Note that these tables show the center frequency for each channel. Many regions impose restrictions on output power and indoor / outdoor use on some channels. Channels are 22MHz wide in 802.442 Channel # 8 9 10 11 12 13 14 Center Frequency (GHz) 2.467 2.417 2. These regulations are rapidly changing.432 2.437 2.452 2.472 2.412 2.11b/g.11a. not all channels are available in all countries.11b / g Channel # 1 2 3 4 5 6 7 Center Frequency (GHz) 2.457 2. 802.484 379 .422 2.

200 5.220 5.745 5.240 5.300 5.11a Channel # 34 36 38 40 42 44 46 48 52 56 60 64 149 153 157 161 Center Frequency (GHz) 5.280 5.380 Appendix B: Channel Allocations 802.805 .190 5.210 5.170 5.180 5.260 5.785 5.320 5.765 5.230 5.

Free Space Path Loss 2400 MHz 5300 MHz Path Loss (dB) 100 110 120 130 140 1 mi 5 km 5 mi 10 km 15 km 10 mi 20 km 25 km 15 mi 30 km 20 mi Distance Appendix C: Path Loss 381 .

002060 0.003276 Max Amperes 302 239 190 150 119 94 75 60 47 37 30 24 19 15 382 .26 2.67 3.11 3. These values can vary from cable to cable.001634 0.62 4. AWG Gauge 0000 000 00 0 1 2 3 4 5 6 7 8 9 10 Diameter (mm) 11.000322 0.Appendix D: Cable Sizes Wire gauge. current capacity.000161 0.000406 0.000815 0.35 6. consult the manufacturer's specifications. diameter.001028 0.27 8.000203 0.002598 0.19 4.25 7.000646 0.91 2.54 5.000256 0. When in doubt.83 5.68 10.40 9.001296 0.59 Ohms / Meter 0. and resistance at 20°C.000513 0.

General Data Site Name Site Latitude (°) Irradiation Data Gdm(0). in kWh / m2 per day) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Worst Irradiation Month Reliability and System Operational Voltage Days of Autonomy (N) Nominal Voltage (VNEquip) 383 .Appendix E: Solar Dimensioning Use these tables to collect the necessary data to estimate the required size of your solar energy system.

384 Appendix E: Solar Dimensioning Component Characteristics Solar Panels Voltage @ Maximum Power (Vpmax) Current @ Maximum Power (Ipmax) Panel Type/Model and Power (Wp) Batteries Nominal Capacity @ 100 H (CNBat) Nominal Voltage (VNBat) Maximum Depth of Discharge (DoDMAX) or Usable Capacity (CUBat) Regulator Nominal Voltage (VNReg) Maximum Current (ImaxReg) DC/AC Inverter (if needed) Nominal Voltage (VNConv) Instantaneous Power (PIConv) Performance @ 70% Load .

Appendix E: Solar Dimensioning 385 Loads Estimated Energy Consumed by the Loads (DC) Month of Greatest Consumption Description # of Units x Nominal Power x Usage Hours / Day = Energy (Wh/day) ETOTAL DC Estimated Energy Consumed by the Loads (AC) Month of Greatest Consumption Description # of Units x Nominal Power x Usage Hours / Day = Energy (Wh/day) ETOTAL AC (before converter) ETOTAL AC (after converter) = ETOTAL AC / 70% .

21 x Im ETOTAL (AC + DC) .386 Appendix E: Solar Dimensioning Finding the Worst Month Site Name Site Latitude (°) Nominal Voltage of the Installation VN (Month) Inclination Gdm (ß) (kWh/m2 J F M A M J J A S O N D ß day) ETOTAL (DC) (Wh/day) ETOTAL (AC) (Wh/day) ETOTAL (AC + DC)= Im (A) = ETOTAL (Wh/day) 1kW/m2/ (Gdm(ß) x VN) Worst Month Summary Worst Month Im (A) ImMAX (A) = 1.

r = 0. mm2/m (for copper) and L is Batteries > Converter Main Line .Vb) For cable thickness computation.Appendix E: Solar Dimensioning 387 Final Calculations Panels Panels in Series (NPS) Panels in Parallel (NPP) Total Number of Panels NPS = VN / VPmax = NPP = ImMAX / IPmax = NTOT = NPS x NPP = Batteries Necessary Capacity (CNEC) Nominal Capacity (CNOM) Number of Batteries in Series (NBS) ETOTAL(WORST MONTH) / VN x N CNEC / DoDMAX VN / VNBAT Cables Panels > Batteries Voltage Drop (Va .Vb) Thickness (Section) r x L x ImMAX / (Va .01286 the length in meters.

.

master mode. A device used to increase the transmitted power of a wireless device. advertised window. Mesh networks often use radios in ad-hoc mode. The portion of a TCP header that specifies how many additional bytes of data the receiver is prepared to accept. address space.11. The distance from the center of a wave to the extreme of one of its peaks. An accounting technique used to manage the expected cost of replacement and obsolescence of equipment over time. An electrical current which varies over time in a cyclic manner. 389 A AC see Alternating Current access point (AP). Alternating Current (AC). While 802. ad-hoc mode. amplifier.11a. master mode accumulator. amplitude. A protocol widely used on Ethernet networks to translate IP addresses into MAC addresses.11 is often used to refer to a family of wireless networking protocols used mainly for local area networking.11b. See also: Direct Current amortization. A radio mode used by 802. See also: CPE. A device that creates a wireless network that is usually connected to a wired Ethernet network.Glossary 0-9 802. 802.11 is a wireless protocol in its own right. AC current is typically used for lighting and appliances. and 802.11g. See also: Wi-Fi.11 devices that allows the creation of a network without an access point. Three popular variants include 802. Another name for a battery. 802. monitor mode . A group of IP addresses that all reside within the same logical subnet. See also: managed mode. Address Resolution Protocol (ARP).

usually by some means of cryptography. AND logic. one-shot execution of programs. anonymizing proxy.390 Glossary coordinates.11 radio is associated to an access point when it is ready to communicate with the network. communications that cannot be linked to a unique individual are said to be anonymous. A logical operation that only evaluates as true if all of the items being compared also evaluate as true. in range of the AP. which means that the effect of gain is present when transmitting as well as receiving. such as through trees. at. Antenna gain is reciprocal.com/argus . anonymity. See also: authenticated antenna diversity.qosient. A graph that describes the relative strength of a radiated field in various directions from an antenna. In computer networks. or other objects. antenna gain. Anonymizing proxies can be used to protect people's privacy and to reduce an organization's exposure to legal liability for the actions of its users. authenticated. and rules about anonymous communications vary widely around the world. A Unix facility that allows timed. polar plot. See also: cron attenuation. An open source network monitoring tool used for tracking flows between hosts. walls. See also: free space loss. A network user that has proven their identity to a service or device (such as an access point) beyond a shadow of a doubt. A technique used to overcome multipath interference by using two or more physically separated receiving antennas. linear polar . using the correct SSID and other authentication parameters. The reduction of available radio power as it is absorbed along a path. The tradeoff of anonymity versus accountability in communications is an ongoing debate online. logarithmic polar coordinates AP see Access Point application layer. Argus see Audit Record Generation and Utilization System ARP see Address Resolution Protocol associated. See also: OR logic. Argus is available from http://www. This means that it is tuned to the proper channel. etc. Business clients of a subscription system who are reliable and can be considered low-risk. See also: rectangular plot. A network service that hides the source or destination of communications. usually expressed in dBi. buildings. scattering Audit Record Generation and Utilization System (Argus). The topmost layer in the OSI and TCP/IP network models. antenna pattern. An 802. See also: anonymity anchor clients. The amount of power concentrated in the direction of strongest radiation of an antenna.

A measure of frequency ranges. both on transmit and receive. A network device that connects two networks together at the data link layer. where the received power is half that of the main lobe. One of several standards used for satellite Internet access.net/ Broadband Global Access Network (BGAN). The angle that measures deviation with respect to the south in the northern hemisphere.net/). throughput battery. inverter beamwidth. The beamwidth of an antenna is usually stated for both the horizontal and vertical planes. A popular web-based monitoring tool written in PHP. load. A feature found on some solar panels that prevents the formation of hot-spots on shaded cells. broadcast address. Benchmarking a network connection typically involves flooding the link with traffic and measuring the actual observed throughput. and with respect to the north in the southern hemisphere. typically used for digital communications. BGAN see Broadband Global Access Network BNC connector. . Bridges do not route packets at the network layer. http://bridge. See also: inclination 391 connectors are typically found on 10base2 coaxial Ethernet. The angular distance between the points on either side of the main lobe of an antenna. On Ethernet networks. bridge-utils. Testing the maximum performance of a service or device. B bandwidth. A coaxial cable connector that uses a "quickconnect" style bayonet lug. benchmarking. See also: solar panel. See also: router and transparent bridging firewall. BNC C CA see Certificate Authority Cacti (http://www. the broadcast IP address is used to send data to all hosts in the local subnet. A Linux software package that is required for creating 802. See also: capacity.1d Ethernet bridges.cacti.sourceforge. They simply repeat packets between two link-local networks. See also: Digital Video Broadcast (DVB-S) and Very Small Aperture Terminal (VSAT). On IP networks. converter. but reduces the maximum voltage of the panel. The word bandwidth is also commonly used interchangeably with capacity to refer to the theoretical maximum data rate of a digital communications line. channel.Glossary azimuth. regulator. A device used to store energy in a photovoltaic system. the broadcast MAC address is used to send data to all machines in the same collision domain. bridge. bypass diodes.

each of which contributes about 2 volts to the battery. often used as a security measure. and automatically change the channel to match it. SSL channel capacity. circular polarization. to display an Acceptable Use Policy). and C networks. which are electrically connected to provide a particular value of current and voltage. CIDR replaces the old classbased addressing scheme. IP address space was allocated in blocks of three different sizes. Certificate Authority. making one full turn for each RF cycle. An electromagnetic field where the electric field vector appears to be rotating with circular motion about the direction of propagation. See also: bandwidth. captive portal. B. Batteries are also made up of individual cells connected in series. An access point that does not broadcast its SSID. these classes are often still referred to and used internally in organizations using private address space. A mechanism used to transparently redirect web browsers to a new location. CIDR was developed to improve routing efficiency on the Internet backbone by enabling route aggregation and network masks of arbitrary size. See also: horizontal polarization. For example.0 can be specified as /24 in CIDR notation.255. cell. The maximum amount of information that can be sent using a given bandwidth. and Class C (255 addresses). Class B (about 65 thousand addresses). These were Class A (about 16 million addresses). 802.392 Glossary capacity.255. B. See also: Class A. A well defined range of frequencies used for communications. and C networks. vertical polarization Class A. the netmask 255. Solar panels are made up of several individual cells. A method used to define a network mask by specifying the number of bits present. See also: Public Key Infrastructure. The theoretical maximum amount of traffic provided by a digital communications line. See also: CIDR notation. For some time. Often used interchangeably with bandwidth. client. Classless Inter-Domain Routing. An 802. throughput. See also: access point. .11 channels use 22 MHz of bandwidth.11 radio card in managed mode. CIDR see Classless Inter-Domain Routing CIDR notation. mesh closed network. Captive portals are often used for authentication or for interrupting a user's online session (for example. See also: Appendix B. While CIDR has replaced class-based allocation. A trusted entity that issues signed cryptographic keys. but are only separated by 5 MHz. data rate channel. Wireless clients will join a network created by an access point.

with the same hardware. On an Ethernet network. outer conductor. The second layer in both the OSI and TCP/IP network models. When collisions are detected. Communications at this layer happen directly between nodes. See also: destructive interference 393 controls. this is also sometimes called the MAC layer. while the maximum throughput is about 20 Mbps). A device used to convert DC signals into a different DC or AC voltage. insulator connectionless protocol. A Unix facility that allows timed and repeated execution of programs. a collision occurs when two devices connected to the same physical segment attempt to transmit at the same time. On Ethernet networks. constructive interference. The speed at which 802. A round (coaxial) cable with a center wire surrounded by a dielectric. See also: inverter CPE see Customer Premises Equipment cron. See also: dielectric. See also: structure converter. Antenna cables are usually made of coax. When two identical waves merge and are in phase. A network protocol (such as UDP) that requires no session initiation or maintenance. the nominal data rate of 802.11g is 54 Mbps. See also: at Customer Premises Equipment. software. and firmware for many components in a network. data rate. Maintenance costs can be reduced by using a consistent platform. conductor. collision. See also: throughput dB see decibel DC see Direct Current DC/AC Converter. A device that converts DC power into AC power. which is always higher than the available throughput. and tough insulating jacket. This is called constructive interference.11 radios exchange symbols. . consistent platform. the amplitude of the resulting wave is twice that of either of the components. Coax is short for "of common axis". See also: session oriented protocol. controls define the RF source in an antenna model. In NEC2. Network equipment (such as a router or bridge) that is installed at a customer's location.Glossary coax. For example. devices delay retransmission for a brief. A material that easily allows electric or thermal energy to flow through without much resistance. randomly selected period. Connectionless protocols typically require less overhead than session oriented protocols. but do not usually offer data protection or packet reassembly. D data link layer.

dipole antenna. See also: Alternating Current Direct Sequence Spread Spectrum (DSSS). default route. Denial of Service (DoS). This is called destructive interference. The ability of an antenna to focus energy in a particular direction when transmitting. The simplest form of omnidirectional antenna. A logarithmic unit of measurement that expresses the magnitude of power relative to a reference level. Digital Elevation Map (DEM). and waveguide antennas. A device that changes the voltage of a DC power source. Digital Video Broadcast (DVB-S). or to receive suitable for use with many appliances. Data that represents the height of terrain for a given geographic area. DC/DC Converter. When a router receives a packet destined for a network for which it has no explicit route. The radio modulation scheme used by 802. A n a c c o u n t i n g method used to save money to cover the eventual break down of equipment. such as access points and routers. A non-conductive material that separates conducting wires inside a cable. See also: Broadband Global Access Network (BGAN) and Very Small Aperture Terminal (VSAT). An attack on network resources. Commonly used units are dBi (decibels relative to an isotropic radiator) and dBm (decibels relative to a milliwatt). The default gateway then repeats the process. the packet is forwarded to the default gateway. usually achieved by flooding a network with traffic or exploiting a bug in an application or network protocol. d e p re c i a t i o n . dish. sectorial antenna directivity.394 Glossary DHCP see Dynamic Host Configuration Protocol dielectric. destructive interference. One of several standards used for satellite Internet access. Also known as an inverter. until the packet reaches its ultimate destination. When two identical waves merge and are exactly out of phase. possibly sending the packet to its own default gateway. A network route that points to the default gateway. An electrical current which remains constant over time. the amplitude of the resulting wave is zero. switching conversion decibel (dB). directional antenna. See also: constructive interference . default gateway. See also: linear conversion. These maps are used by programs such as Radio Mobile to model electromagnetic propagation. Examples of directional antennas include the yagi. Direct Current (DC).11b. DC current is typically used for network equipment. An antenna that radiates very strongly in a particular direction. See also: omnidirectional antenna.

By installing a DNS server on your local LAN. See also: mid span injectors end-to-end encryption. improving response times. An encrypted connection negotiated by both ends of a communications session. dominant mode.uk/ Domain Name Service (DNS). and X rays. dnsmasq.3af Power over Ethernet device that provides power via the Ethernet cable. Someone who intercepts network data such as passwords. electromagnetic spectrum. voice data. The lowest frequency that can be transmitted by a waveguide of a given size. Parts of the electromagnetic spectrum include radio.sourceforge. An 802. A wave that propagates through space without the need for a propagating medium. The place where one organization's network meets another. diversity see antenna diversity DNS see Domain Name Service DNS caching. The widely used network protocol that maps IP addresses to names. which often acts as a firewall. See also: mechanical wave elevation see inclination end span injectors.net/ Ethereal see Wireshark. Dynamic Host Configuration Protocol (DHCP).org. microwave. The very wide range of possible frequencies of electromagnetic energy. This technique is called DNS caching. Edges are defined by the location of the external router. visible light. A protocol used by hosts to automatically determine their IP address.Glossary 395 energy from a particular direction when receiving. DNS requests for an entire network may be cached locally. End-to-end encryption can provide stronger protection than link layer encryption when used on untrusted networks (such as the Internet). An Ethernet switch that provides power on each port is an example of an end span injector. It contains an electric and a magnetic component. available from http://thekelleys. email. edge. . DoS see Denial of Service D S S S s e e D i re c t S e q u e n c e Spread Spectrum DVB-S see Digital Video Broadcast. EtherApe. electromagnetic wave. An open source caching DNS and DHCP server. E eavesdropper. An open source network visualization tool. Available at http://etherape. or online chat.

396

Glossary

Extended Service Set Identifier (ESSID). The name used to identify an 802.11 network. See also: closed network external traffic. Network traffic that originates from, or is destined for, an IP address outside your internal network, such as Internet traffic.

ing loops need to be resolved for proper network operations. free space loss. Power diminished by geometric spreading of the wavefront, as the wave propagates through space. See also: attenuation, free space loss, Appendix C frequency. The number of whole waves that pass a fixed point in a period of time. See also: wavelength, Hertz front-to-back ratio. The ratio of the maximum directivity of an antenna to its directivity in the opposite direction. full duplex. Communications equipment that can send and receive at the same time (such as a telephone). See also: half duplex fwbuilder. A graphical tool that lets you create iptables scripts on a machine separate from your server, and then transfer them to the server later. http://www.fwbuilder.org/

F
firestarter. A graphical front-end for configuring Linux firewalls available from http://www.fs-security.com/. filter. The default table used in the Linux netfilter firewall system is the filter table. This table is used for determining traffic that should be accepted or denied. firewall. A router that accepts or denies traffic based on some criteria. Firewalls are one basic tool used to protect entire networks from undesirable traffic. flush. To remove all entries in a routing table or netfilter chain. forwarding. When routers receive packets that are destined for a different host or network, they send the packet to the next router closest to its ultimate destination. This process is called forwarding. forwarding loops. A routing misconfiguration where packets are forwarded cyclically between two or more routers. Catastrophic network failure is prevented by using the TTL value on every packet, but forward-

G
gain. The ability of a radio component (such as an antenna or amplifier) to increase the power of a signal. See also: decibel gain transfer. Comparing an antenna under test against a known standard antenna, which has a calibrated gain. gasification. The production bubbles of oxygen and hydrogen that

Glossary occurs when a battery is overcharged. globally routable. An address issued by an ISP or RIR that is reachable from any point on the Internet. In IPv4, there are approximately four billion possible IP addresses, although not all of these are globally routable.

397

zontal direction. See also: circular polarization, vertical polarization hot-spot. In wireless networks, a hot-spot is a location that provides Internet access via Wi-Fi, typically by use of a captive portal. In photovoltaic systems, a hot-spot occurs when a single cell in a solar panel is shaded, causing it to act as a resistive load rather than to generate power. hub. An Ethernet networking device that repeats received data on all connected ports. See also: switch. Huygens principle. A wave model that proposes an infinite number of potential wavefronts along every point of an advancing wavefront. Hz see Hertz

H
half duplex. Communications equipment that can send or receive, but never both at once (such as a handheld radio). See also: full duplex. Heliax. High quality coaxial cable that has a solid or tubular center conductor with a corrugated solid outer conductor which enables it to flex. See also: coax Hertz (Hz). A measure of frequency, denoting some number of cycles per second. HF (High-Frequency). Radio waves from 3 to 30 MHz are referred to as HF. Data networks can be built on HF that operate at very long range, but with very low data capacity. hop. Data that crosses one network connection. A web server may be several hops away from your local computer, as packets are forwarded from router to router, eventually reaching their ultimate destination. horizontal polarization. An electromagnetic field with the electric component moving in a linear hori-

I
IANA see Internet Assigned Numbers Authority ICMP see Internet Control Message Protocol ICP see Inter-Cache Protocol impedance. The quotient of voltage over current of a transmission line, consisting of a resistance and a reactance. The load impedance must match the source impedance for maximum power transfer (50 for most communications equipment). inbound traffic. Network packets that originate from outside the local network (typically the Internet) and

398

Glossary Intrusion Detection System (IDS). A program that watches network traffic, looking for suspicious data or behavior patterns. An IDS may make a log entry, notify a network administrator, or take direct action in response to undesirable traffic. inverter see DC/AC Converter IP see Internet Protocol iproute2. The advanced routing tools package for Linux, used for traffic shaping and other advanced techniques. Av a i l a b l e f r o m http://linux-net.osdl.org/ iptables. The primary command used to manipulate netfilter firewall rules. irradiance. The total amount of solar energy that lights a given area, in W/m2 ISM band. ISM is short for Industrial, Scientific, and Medical. The ISM band is a set of radio frequencies set aside by the ITU for unlicensed use. isotropic antenna. A hypothetical antenna that evenly distributes power in all directions, approximated by a dipole. IV characteristic curve. A graph that represents the current that is provided based on the voltage generated for a certain solar radiation.

are bound for a destination inside the local network. See also: outbound traffic. inclination. The angle that marks deviation from a horizontal plane. See also: azimuth infrastructure mode see master mode insulator see dielectric Inter-Cache Protocol (ICP). A high performance protocol used to communicate between web caches. Internet Assigned Numbers Authority (IANA). The organization that administers various critical parts of Internet infrastructure, including IP address allocation, DNS root name servers, and protocol service numbers. Internet Control Message Protocol (ICMP). A Network Layer protocol used to inform nodes about the state of the network. ICMP is part of the Internet protocol suite. See also: Internet protocol suite. Internet layer see network layer Internet Protocol (IP). The most common network layer protocol in use. IP defines the hosts and networks that make up the global Internet. Internet protocol suite (TCP/IP). The family of communication protocols that make up the Internet. Some of these protocols include TCP, IP, ICMP, and UDP. Also called the TCP/IP protocol suite, or simply TCP/IP.

Glossary

399

K
knetfilter. A graphical front-end for configuring Linux firewalls. Available from http://venom.oltrelinux.com/ known good. In troubleshooting, a known good is any component that can be substituted to verify that its counterpart is in good, working condition.

Line of Sight (LOS). If a person standing at point A has an unobstructed view of point B, then point A is said to have a clear Line of Sight to point B. linear polar coordinates. A graph system with equally spaced, graduated concentric circles representing an absolute value on a polar projection. Such graphs are typically used to represent antenna radiation patterns. See also: logarithmic polar coordinates linear conversion. A DC voltage conversion method that lowers the voltage by converting excess energy to heat. See also: switching conversion linear polarization. An electromagnetic wave where the electric field vector stays in the same plane all the time. The electric field may leave the antenna in a vertical orientation, a horizontal orientation, or at some angle between the two. See also: vertical polarization, horizontal polarization link budget. The amount of radio energy available to overcome path losses. If the available link budget exceeds the path loss, minimum receive sensitivity of the receiving radio, and any obstacles, then communications should be possible. link layer encryption. An encrypted connection between link-local devices, typically a wireless client and an access point. See also: end-toend encryption

L
lag. Common term used to describe a network with high latency. lambda ( ) see wavelength LAN see Local Area Network latency. The amount of time it takes for a packet to cross a network connection. It is often (incorrectly) used interchangeably with Round Trip Time (RTT), since measuring the RTT of a wide-area connection is trivial compared to measuring the actual latency. See also: Round Trip Time. lead-acid batteries. Batteries consisting of two submerged lead electrodes in an electrolytic solution of water and sulfuric acid. See also: stationary batteries lease time. In DHCP, IP addresses are assigned for a limited period of time, known as the lease time. After this time period expires, clients must request a new IP address from the DHCP server.

400

Glossary

link-local. Network devices that are connected to the same physical segment communicate with each other directly are said to be linklocal. A link-local connection cannot cross a router boundary without using some kind of encapsulation, such as a tunnel or a VPN. listen. Programs that accept connections on a TCP port are said to listen on that port. load. Equipment in a photovoltaic system that consumes energy. See also: battery, solar panel, regulator, converter, inverter Local Area Network (LAN). A network (typically Ethernet) used within an organization. The part of a network that exists just behind an ISP's router is generally considered to be part of the LAN. See also: WAN. logarithmic polar coordinates. A graph system with logarithmically spaced, graduated concentric circles representing an absolute value on a polar projection. Such graphs are typically used to represent antenna radiation patterns. See also: linear polar coordinates long fat pipe network. A network connection (such as VSAT) that has high capacity and high latency. In order to achieve the best possible performance, TCP/IP must be tuned to match the traffic on such links. LOS see Line of Sight

M
MAC layer see data link layer MAC address. A unique 48 bit number assigned to every networking device when it is manufactured. The MAC address is used for link-local communications. MAC filtering. An access control method based on the MAC address of communicating devices. MAC table. A network switch must keep track of the MAC addresses used on each physical port, in order to efficiently distribute packets. This information is kept in a table called the MAC table. maintenance-free lead-acid batteries see lead-acid batteries Man-In-The-Middle (MITM). A network attack where a malicious user intercepts all communications between a client and a server, allowing information to be copied or manipulated. managed hardware. Networking hardware that provides an administrative interface, port counters, SNMP, or other interactive features is said to be managed. managed mode. A radio mode used by 802.11 devices that allows the radio to join a network created by an access point. See also: master mode, ad-hoc mode, monitor mode

Glossary master browser. On Windows networks, the master browser is the computer that keeps a list of all the computers, shares and printers that are available in Network Neighborhood or My Network Places. master mode. A radio mode used by 802.11 devices that allows the radio to create networks just as an access point does. See also: managed mode, ad-hoc mode, monitor mode match condition. In netfilter, a match condition specifies the criteria that determine the ultimate target for a given packet. Packets may be matched on MAC address, source or destination IP address, port number, data contents, or just about any other property. Maximum Depth of Discharge (DoDmax). The amount of energy extracted from a battery in a single discharge cycle, expressed as a percentage. Maximum Power Point (Pmax). The point where the power supplied by a solar panel is at maximum. MC-Card. A very small microwave connector found on Lucent / Orinoco / Avaya equipment. mechanical wave. A wave caused when some medium or object is swinging in a periodic manner. See also: electromagnetic wave Media Access Control layer see data link layer

401

mesh. A network with no hierarchical organization, where every node on the network carries the traffic of every other as needed. Good mesh network implementations are selfhealing, which means that they automatically detect routing problems and fix them as needed. message types. Rather that port numbers, ICMP traffic uses message types to define the type of information being sent. See also: ICMP. method of the worst month. A method for calculating the dimensions of a standalone photovoltaic system so it will work in the month in which the demand for energy is greatest with respect to the available solar energy. It is the worst month of the year, as this month with have the largest ratio of demanded energy to available energy. MHF see U.FL microfinance. The provision of small loans, savings and other basic financial services to the world s poorest people. mid span injectors. A Power over Ethernet device inserted between an Ethernet switch and the device to be powered. See also: end span injectors milliwatts (mW). A unit of power representing one thousandth of a Watt. MITM see Man-In-The-Middle

402

Glossary

MMCX. A very small microwave connector commonly found on equipment manufactured by Senao and Cisco. monitor mode. A radio mode used by 802.11 devices not normally used for communications that allows the radio passively monitor radio traffic. See also: master mode, managed mode, ad-hoc mode monitor port. On a managed switch, one or more monitor ports may be defined that receive traffic sent to all of the other ports. This allows you to connect a traffic monitor server to the port to observe and analyze traffic patterns. Multi Router Traffic Grapher (MRTG). An open source tool used for graphing traffic statistics. Available from http://oss.oetiker.ch/mrtg/ multipath. The phenomenon of reflections of a signal reaching their target along different paths, and therefore at different times. multipoint-to-multipoint see mesh mW see milliwatt My TraceRoute (mtr). A network diagnostic tool used as an alternative to the traditional traceroute program. http://www.bitwizard.nl/mtr/. See also: traceroute / tracert.

door networking components, such as antennas and outdoor access points. Nagios (http://nagios.org/) A realtime monitoring tool that logs and notifies a system administrator about service and network outages. NAT see Network Address Translation nat. The table used in the Linux netfilter firewall system to configure Network Address Translation. NEC2 see Numerical Electromagnetics Code NetBIOS. A session layer protocol used by Windows networking for file and printer sharing. See also: SMB. netfilter. The packet filtering framework in modern Linux kernels is known as netfilter. It uses the iptables command to manipulate filter rules. http://netfilter.org/ netmask (network mask). A netmask is a 32-bit number that divides the 16 million available IP addresses into smaller chunks, called subnets. All IP networks use IP addresses in combination with netmasks to logically group hosts and networks. NeTraMet. An open source network flow analysis tool available from freshmeat.net/projects/netramet/ network address. The lowest IP number in a subnet. The network address is used in routing tables to specify the destination to be used

N
N connector. A sturdy microwave connector commonly found on out-

Glossary

403

when sending packets to a logical group of IP addresses. Network Address Translation (NAT). NAT is a networking technology that allows many computers to share a single, globally routable IP address. While NAT can help to solve the problem of limited IP address space, it creates a technical challenge for two-way services, such as Voice over IP. network detection. Network diagnostic tools that display information about wireless networks, such as the network name, channel, and encryption method used. network layer. Also called the Internet layer. This is the third layer of the OSI and TCP/IP network models, where IP operates and Internet routing takes place. network mask see netmask ngrep. An open source network security utility used to find patterns in data flows. Available for free from http://ngrep.sourceforge.net/ node. Any device capable of sending and receiving data on a network. Access points, routers, computers, and laptops are all examples of nodes. Nominal Capacity (CN). The maximum amount of energy that can be extracted from a fully charged battery. It is expressed in Ampere-hours (Ah) or Watt-hours (Wh).

Nominal Voltage (VN). The operating voltage of a photovoltaic system, typically 12 or 24 volts. ntop. A network monitoring tool that provides extensive detail about connections and protocol use on a local area network. http://www.ntop.org/ null. In an antenna radiation pattern, a null is a zone in which the effective radiated power is at a minimum. nulling. A specific case of multipath interference where the signal at the receiving antenna is zeroed by the destructive interference of reflected signals. number of days of autonomy (N). The maximum number of days that a photovoltaic system can operate without significant energy received from the sun. Numerical Electromagnetics Code (NEC2). A free antenna modeling package that lets you build an antenna model in 3D, and then analyze the antenna s electromagnetic response. http://www.nec2.org/

O
OFDM see Orthogonal Frequency Division Multiplexing omnidirectional antenna. An antenna that radiates almost equally in every direction in the horizontal plane. See also: directional antenna, sectorial antenna

the electrolyte begins to break down. If energy is applied to a battery beyond its point of maximum charge. passive POE injector see Power over Ethernet . overcharge. A wireless repeater that only uses a single radio. P packet. and protocols. partition. Network packets that originate from the local network and are bound for a destination outside the local network (typically somewhere on the Internet). Hubs will temporarily remove the abusive computer (partition it) from the rest of the network. The OSI model consists of seven interdependent layers. which results in deterioration of the battery. wrapping routing information in a number of encrypted layers. The state of a battery when charge is applied beyond the limit of the battery's capacity. OR logic. Packets are reassembled again at the remote end by TCP (or another protocol) before being passed to the application. A logical operation that evaluates as true if any of the items being compared also evaluate as true. from the physical through the application. at significantly reduced throughput. such as a peer-to-peer client or network virus. A technique used by network hubs to limit the impact of computers that transmit excessively. Excessive partitioning indicates the presence of an excessive bandwidth consumer. Orthogonal Frequency Division Multiplexing (OFDM) OSI network model. Discharging a battery beyond its Maximum Depth of Discharge. On IP networks. one-arm repeater. See also: repeater onion routing. and other routing information that is used to route it to its ultimate destination. and reconnect it again after some time. A privacy tool (such as Tor) that repeatedly bounces your TCP connections across a number of servers spread throughout the Internet. port numbers. destination. To allow more users than the maximum available bandwidth can support. See also: AND logic. Packets are either permitted or discarded depending on the packet filter rules. packet filter. Each packet includes a source. messages sent between computers are broken into small pieces called packets. See also: inbound traffic. oversubscribe. but will remove power before the battery is damaged. See also: TCP/IP network model. A firewall that operates at the Internet layer by inspecting source and destination IP addresses. Regulators will allow a small amount of overcharge time to a battery to avoid gasification. outbound traffic.404 Glossary overdischarge. A popular model of network communications defined by the ISO/IEC 7498-1 standard.

The physical layer is the actual medium used for communications. The classic example of a point-tomultipoint network is an access point at an office with several laptops using it for Internet access. vertical polarization. regulator. or radio waves. . The use of solar panels to collect solar energy to produce electricity. A network protocol typically used on serial lines (such as a dial-up connection) to provide IP connectivity. circular polarization polarization mismatch. Average value of daily irradiation for a given area. multipoint-tomultipoint point-to-point. solar panel. converter. pigtail. Loss of radio signal due to the distance between communicating stations. resulting in signal loss. See also: point-to-multipoint. See also: horizontal polarization. Ping can be used to determine the location of network problems by "pinging" computers in the path between the local machine and the ultimate destination. A standalone photovoltaic system does this without any connection to an established power grid. such as copper cable. A state where a transmitting and receiving antenna do not use the same polarization.Glossary path loss. The direction of the electric component of an electromagnetic wave as it leaves the transmitting antenna. A ubiquitous network diagnostic utility that uses ICMP echo request and reply messages to determine the round trip time to a network host. polar plot. See also: thermal solar energy photovoltaic system. See also: point-to-point. An energy system that generates electrical energy from solar radiation and stores it for later use. The lowest layer in both the OSI and TCP/IP network models. A graph where points are located by projection along a rotating axis (radius) to an intersection with one of several concentric circles. A heavy piece of metal buried in the earth to improve a ground connection. PoE see Power over Ethernet point-to-multipoint. optic fiber. A wireless network where several nodes connect back to a central location. load. See also: battery. Peak Sun Hours (PSH). A short microwave cable that converts a non-standard connector into something more robust and commonly available. 405 PKI see Public Key Infrastructure plomb. See also: rectangular plot polarization. ping. photovoltaic generator see solar panel photovoltaic solar energy. A wireless network consisting of only two stations. inverter physical layer. usually separated by a great distance. multipoint-to-multipoint Point-to-Point Protocol (PPP).

A diagnostic program used to observe and disassemble network packets.0/16. private address space. A form of encryption used by SSL. protocol stack. SSH. Privoxy (http://www. and 192. as well as errors and retransmissions.0/12.16. See also: end span injectors.0/8. the default policy for any chain may be set to ACCEPT or DROP. Public Key Infrastructure (PKI). These statistics may include inbound and outbound packet and byte counts. PSH see Peak Sun Hours Public key cryptography.0. 172. the policy is the default action to be taken when no other filtering rules apply.privoxy. proactive routing.168. In netfilter.0. The sixth layer of the OSI networking model. The amount of energy in a certain amount of time. See also: reactive routing protocol analyzer.0. A web proxy that provides anonymity through the use of filters. such as MIME encoding or data compression. A set of network protocols that provide interdependent layers of functionality. A set of reserved IP addresses outlined in RFC1918. Private address space is frequently used within an organization. Privoxy is often used in conjunction with Tor. Power over Ethernet (PoE). in conjunction with Network Address Translation (NAT). The reserved private address space ranges include 10. A mesh implementation where every node knows about the existence of every other node in the mesh cloud as well as which nodes may be used to route traffic to them.org/). Protocol analyzers provide the greatest possible detail about individual packets. For example. Managed switches and routers provide statistics for each network port called port counters.0. mid span injectors PPP see Point to Point Protocol presentation layer. Public key cryptography allows encrypted information to be exchanged over an untrusted network without the need to distribute a secret key. and other popular security programs. See also: certificate authority . A technique used to supply DC power to devices using the Ethernet data cable.406 Glossary policy. A security mechanism used in conjunction with public key cryptography to prevent the possibility of Man-In-The-Middle attacks. This layer deals with data representation. See also: OSI network model and TCP/IP network model. See also: NAT. port counters. Each node maintains a routing table covering the whole mesh cloud. power.

A network monitoring tool that performs unattended monitoring over long periods. The portion of the electromagnetic spectrum in which waves can be generated by applying alternating current to an antenna.net/. reciprocity. See also: impedance reverse polarity (RP). A graph where points are located on a simple grid. or eventually become de facto standards. The component of a photovoltaic system that assures that the battery is working in appropriate conditions. recombinant batteries see leadacid batteries rectangular plot. A type of fuse that immediately blows if the current flowing through it is higher than their rating. See also: solar panel. A node that is configured to rebroadcast traffic that is not destined for the node itself. both of which are very detrimental to the life of the battery. which are delegated to one of the five regional Internet registries. See also: proactive routing realtime monitoring. An antenna's ability to maintain the same characteristics regardless if whether it is transmitting or receiving. Proprietary microwave connectors. load. The 4 billion available IP ad- . A logarithmic ratio measured in dB that compares the power reflected by the antenna to the power that is fed into the antenna from the transmission line. See also: slow blow dresses are administered by the IANA. return loss. inverter repeater.Glossary 407 Q quick blow. It avoids overcharging or undercharging the battery. battery. and notifies administrators immediately when problems arise. but many are either approved explicitly by the IETF. RFCs can be viewed online at http://rfc. The space has been divided into large subnets. each with authority over a large geographic area. RFCs are a numbered series of documents published by the Internet Society that document ideas and concepts related to Internet technologies. Not all RFCs are actual standards. radio. reactive routing. regulator. based on a standard connector but with the genders reversed. A mesh implementation where routes are computed only when it is necessary to send data to a specific node. converter. See also: polar plot Regional Internet Registrars (RIR). often used to extend the useful range of a network. Request for Comments (RFC). The RP-TNC is R radiation pattern see antenna pattern.

but others (such as RP-SMA and RP-N) are also commonplace.samba. This is the data format used by RRDtool and other network monitoring tools. or by a malicious person who intends to collect data or do harm to the network. rogue access points.oetiker. RF transmission line. the router uses its default gateway.ch/rrdtool/ rsync (http://rsync. Frequently confused with latency. A database that stores information in a very compact way that does not expand over time. router. RRDtool is available from http://oss. as well as generate useful graphs to present the data. The process of forwarding packets to the next hop is called routing. An unauthorized access point incorrectly installed by legitimate users. RTT see Round Trip Time S SACK see Selective Acknowledgment scattering. or server load average) and can display that data as an average over time. A suite of tools that allow you to create and modify RRD databases. routing. Routers operate at the Network Layer. A device that forwards packets between different networks. The amount of time it takes for a packet to be acknowledged from the remote end of a connection. See also: bridge and default gateway. Round Robin Database (RRD). RRD see Round Robin Database RRDtool. with the genders reversed.408 Glossary probably the most common reverse polarity connector. or a waveguide) between a radio and an antenna. See also: free space loss. attenuation . A list of networks and IP addresses kept by a router to determine how packets should be forwarded.org/). RRDtool is used to keep track of time-series data (such as network bandwidth. The RP-TNC is often found on equipment manufactured by Linksys. A device that does this is called a router. Heliax. An open source incremental file transfer utility used for maintaining mirrors. RIR see Regional Internet Registrars Round Trip Time (RTT). machine room temperature. The connection (typically coax. Signal loss due to objects in the path between two nodes. The process of forwarding packets between different networks. routing table. If a router receives a packet for a network that is not in the routing table. A common proprietary version of the TNC microwave connector. RP see Reverse Polarity RP-TNC.

while connectionless protocols do not. . sidelobes. Layer five of the OSI model. Simple Network Management Protocol (SNMP).Glossary sectorial antenna. Some is inevitably radiated in other directions. you are using SSL. A mechanism used to overcome TCP inefficiencies on high latency networks. A transmitter that emits continuously at a specific frequency. slow blow. SSL uses public key cryptography and a trusted public key infrastructure to secure data communications on the web. Selective Acknowledgment (SACK). Whenever you visit a web URL that starts with https. A network protocol (such as TCP) that requires initialization before data can be exchanged. A configuration tool used for setting up netfilter firewalls without the need to learn iptables syntax. A network protocol used in Windows networks to provide file sharing services. See also: directional antenna. Session oriented protocols typically offer error correction and packet reassembly. An antenna that radiates primarily in a specific area. A link-local network where every node can observe the traffic of every other node. A small threaded microwave connector. signal generator. See also: connectionless protocol. A fuse that allows a current higher than its rating to pass for a short time. 409 shared medium. See also: Squid. SNMP is typically used to poll network switches and routers to gather operating statistics. large organizations can improve efficiency by installing a site-wide web cache. See also: NetBIOS. such as Squid. A site-wide web cache keeps a copy of all requests made from within an organization. A protocol designed to facilitate the exchange of management information between network devices. as well as some clean-up after data exchange has completed. site-wide web cache. These smaller peaks are referred to as sidelobes. session oriented protocol. and serves the local copy on subsequent requests. omnidirectional antenna Secure Sockets Layer (SSL). Shorewall (http://shorewall. No antenna is able to radiate all the energy in one preferred direction. or as narrow as 60 degrees. such as VSAT. Server Message Block (SMB). See also: quick blow SMA. the Session Layer manages logical connections between applications. An end-to-end encryption technology built into virtually all web browsers. While all modern web browsers provide a local data cache. Service Set ID (SSID) see Extended Service Set Identifier session layer.net/). The beam can be as wide as 180 degrees.

. converter. A latency measurement tool that measures. The state is not part of the packet as transmitted over the Internet. but is determined by the firewall itself. latency distribution and packet loss all on a single graph. or service. full-featured. See also: Intrusion Detection System. load. The component of a photovoltaic system used to convert solar radiation into electricity. user. solar power charge regulator see regulator spectrum see electromagnetic spectrum spectrum analyzer. A very popular open source web proxy cache. capacity. See also: bandwidth. A technique used to serve different answers to DNS requests based on the source of the request. stateful inspection. SoC see State of Charge solar module see solar panel solar panel.snort. Network monitoring tools that are run only when needed to diagnose a problem. spoof. and latency. http://www. regulator. A "high-speed" network should have low latency and more than enough capacity to carry the traffic of its users. See also: Wi-Spy Speed. robust.oetiker. A device that provides a visual representation of the electromagnetic spectrum. spot check tools. Firewall rules that are aware of the the state associated with a given packet. split horizon DNS.410 SMB see Server Message Block Glossary SmokePing. To impersonate a network device. determined by the current voltage and type of battery.squid-cache. New.org/ SSID see Extended Service Set Identifier SSL see Secure Sockets Layer standalone photovoltaic system see photovoltaic system State of Charge (SoC).ch/smokeping/ SNMP see Simple Network Management Protocol Snort (http://www. Squid.org/). Ping and traceroute are examples of spot check tools. See also: battery. stores and displays latency. The amount of charge present in a battery. inverter solar panel array. and scales to support networks of nearly any size. A generic term used to refer to the responsiveness of a network connection. A set of solar panels wired in series and/or parallel in order to provide the necessary energy for a given load. It is flexible. Split horizon is used to direct internal users to a different set of servers than Internet users. A very popular open source intrusion detection system. SmokePing is available from http://oss.

A DC voltage conversion method that uses a magnetic component to temporarily store the energy and transform it to another voltage. A popular simplification of the OSI network model that is used with Internet networks. Stateful inspection is sometimes called connection tracking. T target. stationary batteries. Batteries designed to have a fixed location and in scenarios where the power consumption is more or less irregular. Temporal Key Integrity Protocol (TKIP). See also: WinDump and Wireshark. vides a nection devices. defined by netmasks. a window size of 3000 would mean that two packets of 1500 bytes each will be sent. the action to be taken once a packet matches a rule. tcpdump.org/. See also: controls subnet mask see netmask subnets. and related connections may all be taken into consideration when filtering packets. See also: OSI network model. Energy collected from the sun in the form of heat. In netfilter. and REJECT. See also: lead-acid batteries structure. DROP. clude ACCEPT. but they are not designed to produce high currents in brief periods of time. An encryption protocol used in conjunction with WPA to improve the security of a communications session. See also: photovoltaic solar energy switching conversion. switch. A popular open source packet capture and analysis tool available at http://www. LOG. The TCP parameter that defines how much data that may be sent before an ACK packet is returned from the receiving side. thermal solar energy. Stationary batteries can accommodate deep discharge cycles. Some possible netfilter targets in- . In NEC2. dedicated conbetween communicating See also: hub. Switching conversion is much more efficient than linear conversion. A subset of a range of IP networks. The TCP/IP model consists of five interdependent layers. TCP/IP see Internet protocol suite TCP/IP network model.tcpdump. A network device that protemporary. from the physical through the application. after which the receiving end will either ACK the chunk or request retransmission.Glossary 411 established. TCP see Transmission Control Protocol TCP acknowledgment spoofing TCP window size. and how the wires are connected up. a numerical description of where the different parts of the antenna are located. For instance.

The amount of power provided by the radio transmitter. before any antenna gain or line losses. The actual amount of information per second flowing through a network connection. A session oriented protocol that operates at the Transport Layer. The state when a computer has exhausted the available RAM and must use the hard disk for temporary storage. In TCP/IP networks. the TTL is a counter that starts at some value (such as 64) and is decremented at each router hop. In DNS. Web requests are silently redirected to the cache. See also: bridge. and also display latency statistics. Tor (http://www.torproject. disregarding protocol overhead. which thrashing. transmission power. See also: mtr. sturdy threaded microwave connector. A method of implementing a site-wide web cache that requires no configuration on the web clients. congestion avoidance. Both use ICMP echo requests with increasing TTL values to determine which routers are used to connect to a remote host. TKIP see Temporal Key Integrity Protocol TNC connector. Time To Live (TTL). If the TTL reaches zero. A ubiquitous network diagnostic utility often used in conjunction with ping to determine the location of network problems. throughput testing tools. including HTTP and SMTP. One benefit of a transparent bridging firewall is that it does not require an IP address. while the Windows version is tracert.412 Glossary traceroute / tracert. The Unix version is called traceroute. This mechanism helps reduce damage caused by routing loops. traction batteries see lead-acid batteries Transmission Control Protocol (TCP). A common. throughput.org/). greatly reducing system performance. transparent cache. the TTL defines how long a cached object may be kept before it must be again retrieved from the original website. transparent bridging firewall. In Squid. the TTL defines the amount of time that a particular zone record should be kept before it must be refreshed. TCP is an integral protocol used by many Internet applications. Tools that measure the actual bandwidth available between two points on a network. See also: UDP. Another variant is tracepath. and reliable delivery. A TTL value acts as a deadline or emergency brake to signal a time when the data should be discarded. An onion routing tool that provides good protection against traffic analysis. the packet is discarded. A firewall technique that introduces a bridge that selectively forwards packets based on firewall rules. . providing packet reassembly. which uses a similar technique with UDP packets.

transport layer. See also: circular polarization.FL. Most wireless consumer electronic devices use vertical polarization. The usable capacity of a battery. The third layer of the OSI and TCP/IP network models. trending. UTP see Unshielded Twisted Pair V valve regulated lead acid battery (VRLA) see lead-acid batteries vertical polarization. Laptop users who accidentally associate to the wrong wireless network.Glossary 413 makes the request on behalf of the client. This is often used in conjunction with encryption to protect communications from potential eavesdroppers. User Datagram Protocol (UDP). See also: site-wide web cache. which helps you plan for upgrades and changes. transparent proxy. while eliminating the need to support encryption within the application itself. Squid. equal to the product of the Nominal Capacity and the Maximum Depth of Discharge. which provides a method of reaching a particular service on a given network node. Transparent caches cannot use authentication. Useful Capacity (Cu ). consisting of four pairs of twisted wires. UDP see User Datagram Protocol unintentional users. vertical polarization . An electromagnetic field with the electric component moving in a linear vertical direction. which makes it impossible to implement traffic accounting at the user level. Cable used for 10baseT and 100baseT Ethernet. A type of network monitoring tool that performs unattended monitoring over long periods. Unshielded Twisted Pair (UTP). A caching proxy installed so that users web requests are automatically forwarded to the proxy server. A very tiny microwave connector commonly used on mini-PCI radio cards. Trending tools allow you to predict future behavior of your network. A form of data encapsulation that wraps one protocol stack within another. TTL see Time To Live tunnel. U U. and plots the results on a graph. A connectionless protocol (at the transport layer) commonly used for video and audio streaming. Examples of protocols that operate at this layer are TCP and UDP. Tunnels are often used conjunction with VPNs. without any need to manually configure web browsers to use it.

414 Glossary See also: Broadband Global Access Network (BGN) and Digital Video Broadcast (DVB-S). Wide Area Network (WAN).gnu.net/. 802. VSAT is the most widely deployed satellite technology used in Africa. One of several standards used for satellite Internet access. VPN see Virtual Private Network. DSL. VSAT is the most widely deployed satellite technology used in Africa. Gizmo Project. wavelength. Wi-Spy. Virtual Private Network (VPN).11g). An inexpensive 2. See also: Broadband Global Access Network (BGAN) and Digital Video Broadcast (DVB-S). for example from the top of one peak to the next. Wireless enthusiasts who are interested in finding the physical location of wireless networks. fixed wireless. and satellite all . video sender. WEP see Wired Equivalent Privacy wget. Wi-Fi is short for Wireless Fidelity. W WAN see Wide Area Network War drivers. frame relay. The distance measured from a point on one wave to the equivalent part of the next.org/software/wget/ Wi-Fi. MSN Messenger. Examples of popular VoIP clients include Skype. http://www. VoIP (Voice over IP). One of several standards used for satellite Internet access. A fairly strong link layer encryption protocol supported by most modern Wi-Fi equipment. A marketing brand owned by the Wi-Fi Alliance that is used to refer to various wireless networking technologies (including 802. and 802.4 GHz spectrum analysis tool available from http://www. Very Small Aperture Terminal (VSAT). VPNs use a combination of encryption and tunneling to secure all network traffic.4 GHz video transmitter that can be used as an inexpensive signal generator. A 2.11a.11b. Leased lines. regardless of the application being used. Also known as lambda ( ). VRLA see valve regulated lead acid battery VSAT see Very Small Aperture Terminal Very Small Aperture Terminal (VSAT). Any long distance networking technology. Wi-Fi Protected Access (WPA).metageek. See also: tunnel. VPNs are often used to connect remote users to an organization's network when traveling or working from home. An open source command line tool for downloading web pages. A tool used to join two networks together over an untrusted network (such as the Internet). and iChat. A technology that provides telephone-like features over an Internet connection.

One of the most popular public wikis is http://www.org/windump/ Wired Equivalent Privacy (WEP).wireshark. It is available from http://www.Glossary 415 typically implement wide area networks.winpcap.zabbix. .org/ window scale.wikipedia. A TCP enhancement defined by RFC1323 that allows TCP window sizes larger than 64KB.org/) A realtime monitoring tool that logs and notifies a system administrator about service and network outages.11a/b/g equipment. wiki. See also: LAN. A somewhat secure link layer encryption protocol supported by virtually all 802. Wireless Fidelity see Wi-Fi. The Windows version of tcpdump. http://www. WinDump. wireshark. A free network protocol analyzer for Unix and Windows.org/ WPA see Wi-Fi Protected Access Z Zabbix (http://www. A web site that allows any user to edit the contents of any page.

Sign up to vote on this title
UsefulNot useful